to 0.2.9

Fix type error when setting encoder_feature in simul_whisper->infer for faster whisper encoder

.gitignore

@@ -137,4 +137,5 @@ run_*.sh
 test_*.py
 launch.json
 .DS_Store
-test/*
+test/*
+nllb-200-distilled-600M-ctranslate2/*

DEV_NOTES.md

@@ -0,0 +1,70 @@
+# 1. Simulstreaming: Decouple the encoder for faster inference
+
+Simulstreaming encoder time (whisperlivekit/simul_whisper/simul_whisper.py l. 397) experimentations :
+
+On macOS Apple Silicon M4 :
+
+| Encoder | base.en | small |
+|--------|---------|-------|
+| WHISPER (no modification) | 0.35s | 1.09s |
+| FASTER_WHISPER | 0.4s | 1.20s |
+| MLX_WHISPER | 0.07s | 0.20s |
+
+Memory saved by only loading encoder for optimized framework:
+
+For tiny.en, mlx whisper:
+Sizes MLX whisper:
+Decoder weights: 59110771 bytes
+Encoder weights: 15268874 bytes
+
+
+
+# 2. SortFormer Diarization: 4-to-2 Speaker Constraint Algorithm
+
+Transform a diarization model that predicts up to 4 speakers into one that predicts up to 2 speakers by mapping the output predictions.
+
+## Problem Statement
+- Input: `self.total_preds` with shape `(x, x, 4)` - predictions for 4 speakers
+- Output: Constrained predictions with shape `(x, x, 2)` - predictions for 2 speakers
+
+#
+### Initial Setup
+For each time step `i`, we have a ranking of 4 speaker predictions (1-4). When only 2 speakers are present, the model will have close predictions for the 2 active speaker positions.
+
+Instead of `np.argmax(preds_np, axis=1)`, we take the top 2 predictions and build a dynamic 4→2 mapping that can evolve over time.
+
+### Algorithm
+
+```python
+top_2_speakers = np.argsort(preds_np, axis=1)[:, -2:]
+```
+
+- `DS_a_{i}`: Top detected speaker for prediction i
+- `DS_b_{i}`: Second detected speaker for prediction i  
+- `AS_{i}`: Attributed speaker for prediction i
+- `GTS_A`: Ground truth speaker A
+- `GTS_B`: Ground truth speaker B
+- `DIST(a, b)`: Distance between detected speakers a and b
+
+3. **Attribution Logic**
+
+```
+AS_0 ← A
+
+AS_1 ← B
+
+IF DIST(DS_a_0, DS_a_1) < DIST(DS_a_0, DS_a_2) AND 
+    DIST(DS_a_0, DS_a_1) < DIST(DS_a_1, DS_a_2):
+    # Likely that DS_a_0 = DS_a_1 (same speaker)
+    AS_1 ← A
+    AS_2 ← B
+
+ELIF DIST(DS_a_0, DS_a_2) < DIST(DS_a_0, DS_a_1) AND 
+    DIST(DS_a_0, DS_a_2) < DIST(DS_a_1, DS_a_2):
+    AS_2 ← A
+
+ELSE:
+    AS_2 ← B
+
+to finish
+```

Dockerfile

@@ -1,4 +1,4 @@
-FROM nvidia/cuda:12.8.1-cudnn-runtime-ubuntu22.04
+FROM nvidia/cuda:12.9.1-cudnn-devel-ubuntu24.04
 
 ENV DEBIAN_FRONTEND=noninteractive
 ENV PYTHONUNBUFFERED=1
@@ -9,48 +9,50 @@ ARG EXTRAS
 ARG HF_PRECACHE_DIR
 ARG HF_TKN_FILE
 
-# Install system dependencies
-#RUN apt-get update && \
-#    apt-get install -y ffmpeg git && \
-#    apt-get clean && \
-#    rm -rf /var/lib/apt/lists/*
-
-# 2) Install system dependencies + Python + pip
 RUN apt-get update && \
     apt-get install -y --no-install-recommends \
         python3 \
         python3-pip \
+        python3-venv \
         ffmpeg \
         git \
         build-essential \
-        python3-dev && \
+        python3-dev \
+        ca-certificates && \
     rm -rf /var/lib/apt/lists/*
 
-RUN pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cu128
+RUN python3 -m venv /opt/venv
+ENV PATH="/opt/venv/bin:$PATH"
+
+# timeout/retries for large torch wheels
+RUN pip3 install --upgrade pip setuptools wheel && \
+    pip3 --disable-pip-version-check install --timeout=120 --retries=5 \
+        --index-url https://download.pytorch.org/whl/cu129 \
+        torch torchaudio \
+    || (echo "Initial install failed — retrying with extended timeout..." && \
+        pip3 --disable-pip-version-check install --timeout=300 --retries=3 \
+            --index-url https://download.pytorch.org/whl/cu129 \
+            torch torchvision torchaudio)
 
 COPY . .
 
 # Install WhisperLiveKit directly, allowing for optional dependencies
-#   Note: For gates models, need to add your HF toke. See README.md
-#         for more details.
 RUN if [ -n "$EXTRAS" ]; then \
       echo "Installing with extras: [$EXTRAS]"; \
-      pip install --no-cache-dir .[$EXTRAS]; \
+      pip install --no-cache-dir whisperlivekit[$EXTRAS]; \
     else \
       echo "Installing base package only"; \
-      pip install --no-cache-dir .; \
+      pip install --no-cache-dir whisperlivekit; \
     fi
 
-# Enable in-container caching for Hugging Face models by: 
+# In-container caching for Hugging Face models by: 
-# Note: If running multiple containers, better to map a shared
-# bucket. 
-#
 # A) Make the cache directory persistent via an anonymous volume.
 #    Note: This only persists for a single, named container. This is 
 #          only for convenience at de/test stage. 
 #          For prod, it is better to use a named volume via host mount/k8s.
 VOLUME ["/root/.cache/huggingface/hub"]
 
+
 # or
 # B) Conditionally copy a local pre-cache from the build context to the 
 #    container's cache via the HF_PRECACHE_DIR build-arg.
@@ -65,8 +67,7 @@ RUN if [ -n "$HF_PRECACHE_DIR" ]; then \
       echo "No local Hugging Face cache specified, skipping copy"; \
     fi
 
-# Conditionally copy a Hugging Face token if provided
+# Conditionally copy a Hugging Face token if provided. Useful for Diart backend (pyannote audio models)
-
 RUN if [ -n "$HF_TKN_FILE" ]; then \
       echo "Copying Hugging Face token from $HF_TKN_FILE"; \
       mkdir -p /root/.cache/huggingface && \
@@ -74,11 +75,9 @@ RUN if [ -n "$HF_TKN_FILE" ]; then \
     else \
       echo "No Hugging Face token file specified, skipping token setup"; \
     fi
-    
+
-# Expose port for the transcription server
 EXPOSE 8000
 
 ENTRYPOINT ["whisperlivekit-server", "--host", "0.0.0.0"]
 
-# Default args
+CMD ["--model", "medium"]
-CMD ["--model", "base"]

Dockerfile.cpu

@@ -0,0 +1,61 @@
+FROM python:3.13-slim
+
+ENV DEBIAN_FRONTEND=noninteractive
+ENV PYTHONUNBUFFERED=1
+
+WORKDIR /app
+
+ARG EXTRAS
+ARG HF_PRECACHE_DIR
+ARG HF_TKN_FILE
+
+RUN apt-get update && \
+    apt-get install -y --no-install-recommends \
+        ffmpeg \
+        git \
+        build-essential \
+        python3-dev && \
+    rm -rf /var/lib/apt/lists/*
+
+# Install CPU-only PyTorch
+RUN pip install torch torchvision torchaudio --index-url https://download.pytorch.org/whl/cpu
+
+COPY . .
+
+# Install WhisperLiveKit directly, allowing for optional dependencies
+RUN if [ -n "$EXTRAS" ]; then \
+      echo "Installing with extras: [$EXTRAS]"; \
+      pip install --no-cache-dir whisperlivekit[$EXTRAS]; \
+    else \
+      echo "Installing base package only"; \
+      pip install --no-cache-dir whisperlivekit; \
+    fi
+
+# Enable in-container caching for Hugging Face models
+VOLUME ["/root/.cache/huggingface/hub"]
+
+# Conditionally copy a local pre-cache from the build context
+RUN if [ -n "$HF_PRECACHE_DIR" ]; then \
+      echo "Copying Hugging Face cache from $HF_PRECACHE_DIR"; \
+      mkdir -p /root/.cache/huggingface/hub && \
+      cp -r $HF_PRECACHE_DIR/* /root/.cache/huggingface/hub; \
+    else \
+      echo "No local Hugging Face cache specified, skipping copy"; \
+    fi
+
+# Conditionally copy a Hugging Face token if provided
+RUN if [ -n "$HF_TKN_FILE" ]; then \
+      echo "Copying Hugging Face token from $HF_TKN_FILE"; \
+      mkdir -p /root/.cache/huggingface && \
+      cp $HF_TKN_FILE /root/.cache/huggingface/token; \
+    else \
+      echo "No Hugging Face token file specified, skipping token setup"; \
+    fi
+    
+# Expose port for the transcription server
+EXPOSE 8000
+
+ENTRYPOINT ["whisperlivekit-server", "--host", "0.0.0.0"]
+
+# Default args - you might want to use a smaller model for CPU
+CMD ["--model", "tiny"]

README.md

@@ -8,8 +8,8 @@
 
 <p align="center">
 <a href="https://pypi.org/project/whisperlivekit/"><img alt="PyPI Version" src="https://img.shields.io/pypi/v/whisperlivekit?color=g"></a>
-<a href="https://pepy.tech/project/whisperlivekit"><img alt="PyPI Downloads" src="https://static.pepy.tech/personalized-badge/whisperlivekit?period=total&units=international_system&left_color=grey&right_color=brightgreen&left_text=downloads"></a>
+<a href="https://pepy.tech/project/whisperlivekit"><img alt="PyPI Downloads" src="https://static.pepy.tech/personalized-badge/whisperlivekit?period=total&units=international_system&left_color=grey&right_color=brightgreen&left_text=installations"></a>
-<a href="https://pypi.org/project/whisperlivekit/"><img alt="Python Versions" src="https://img.shields.io/badge/python-3.9--3.13-dark_green"></a>
+<a href="https://pypi.org/project/whisperlivekit/"><img alt="Python Versions" src="https://img.shields.io/badge/python-3.9--3.15-dark_green"></a>
 <a href="https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/LICENSE"><img alt="License" src="https://img.shields.io/badge/License-MIT/Dual Licensed-dark_green"></a>
 </p>
 
@@ -19,6 +19,7 @@ Real-time speech transcription directly to your browser, with a ready-to-use bac
 #### Powered by Leading Research:
 
 - [SimulStreaming](https://github.com/ufal/SimulStreaming) (SOTA 2025) - Ultra-low latency transcription with AlignAtt policy
+- [NLLB](https://arxiv.org/abs/2207.04672), ([distilled](https://huggingface.co/entai2965/nllb-200-distilled-600M-ctranslate2)) (2024) - Translation to more than 100 languages.
 - [WhisperStreaming](https://github.com/ufal/whisper_streaming) (SOTA 2023) - Low latency transcription with LocalAgreement policy
 - [Streaming Sortformer](https://arxiv.org/abs/2507.18446) (SOTA 2025) - Advanced real-time speaker diarization
 - [Diart](https://github.com/juanmc2005/diart) (SOTA 2021) - Real-time speaker diarization
@@ -39,6 +40,8 @@ Real-time speech transcription directly to your browser, with a ready-to-use bac
 ```bash
 pip install whisperlivekit
 ```
+> You can also clone the repo and `pip install -e .` for the latest version.
+
 
 >  **FFmpeg is required** and must be installed before using WhisperLiveKit
 > 
@@ -66,41 +69,32 @@ pip install whisperlivekit
 
 | Optional | `pip install` |
 |-----------|-------------|
-| Speaker diarization | `whisperlivekit[diarization]` |
+| **Speaker diarization with Sortformer** | `git+https://github.com/NVIDIA/NeMo.git@main#egg=nemo_toolkit[asr]` |
-| Original Whisper backend | `whisperlivekit[whisper]` |
+| **Apple Silicon optimized backend** | `mlx-whisper` |
-| Improved timestamps backend | `whisperlivekit[whisper-timestamped]` |
+| **NLLB Translation** | `huggingface_hub` & `transformers` |
-| Apple Silicon optimization backend | `whisperlivekit[mlx-whisper]` |
+| *[Not recommanded]*  Speaker diarization with Diart | `diart` |
-| OpenAI API backend | `whisperlivekit[openai]` |
+| *[Not recommanded]*  Original Whisper backend | `whisper` |
+| *[Not recommanded]*  Improved timestamps backend | `whisper-timestamped` |
+| OpenAI API backend | `openai` |
 
 See  **Parameters & Configuration** below on how to use them.
 
- 
-> **Pyannote Models Setup** For diarization, you need access to pyannote.audio models:
-> 1. [Accept user conditions](https://huggingface.co/pyannote/segmentation) for the `pyannote/segmentation` model
-> 2. [Accept user conditions](https://huggingface.co/pyannote/segmentation-3.0) for the `pyannote/segmentation-3.0` model
-> 3. [Accept user conditions](https://huggingface.co/pyannote/embedding) for the `pyannote/embedding` model
->4. Login with HuggingFace:
-> ```bash
-> huggingface-cli login
-> ```
 
-## 💻 Usage Examples
 
-#### Command-line Interface
+### Usage Examples
 
-Start the transcription server with various options:
+**Command-line Interface**: Start the transcription server with various options:
 
 ```bash
-# SimulStreaming backend for ultra-low latency
+# Use better model than default (small)
-whisperlivekit-server --backend simulstreaming --model large-v3
+whisperlivekit-server --model large-v3
 
-# Advanced configuration with diarization
+# Advanced configuration with diarization and language
 whisperlivekit-server --host 0.0.0.0 --port 8000 --model medium --diarization --language fr
 ```
 
 
-#### Python API Integration (Backend)
+**Python API Integration**: Check [basic_server](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/whisperlivekit/basic_server.py) for a more complete example of how to use the functions and classes.
-Check [basic_server](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/whisperlivekit/basic_server.py) for a more complete example of how to use the functions and classes.
 
 ```python
 from whisperlivekit import TranscriptionEngine, AudioProcessor, parse_args
@@ -138,18 +132,29 @@ async def websocket_endpoint(websocket: WebSocket):
         await audio_processor.process_audio(message)        
 ```
 
-#### Frontend Implementation
+**Frontend Implementation**: The package includes an HTML/JavaScript implementation [here](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/whisperlivekit/web/live_transcription.html). You can also import it using `from whisperlivekit import get_inline_ui_html` & `page = get_inline_ui_html()`
-
-The package includes an HTML/JavaScript implementation [here](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/whisperlivekit/web/live_transcription.html). You can also import it using `from whisperlivekit import get_web_interface_html` & `page = get_web_interface_html()`
 
 
-### ⚙️ Parameters & Configuration
+## Parameters & Configuration
+
+An important list of parameters can be changed. But what *should* you change?
+- the `--model` size. List and recommandations [here](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/available_models.md)
+- the `--language`.  List [here](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/whisperlivekit/simul_whisper/whisper/tokenizer.py). If you use `auto`, the model attempts to detect the language automatically, but it tends to bias towards English.
+- the `--backend` ? you can switch to `--backend faster-whisper` if  `simulstreaming` does not work correctly or if you prefer to avoid the dual-license requirements.
+- `--warmup-file`, if you have one
+- `--task translate`, to translate in english
+- `--host`, `--port`, `--ssl-certfile`, `--ssl-keyfile`, if you set up a server
+- `--diarization`, if you want to use it.
+- [BETA] `--target-language`, to translate using NLLB. [118 languages available](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/whisperlivekit/translation/mapping_languages.py). If you want to translate to english, you should rather use `--task translate`, since Whisper can do it directly.
+
+### Full list of parameters :
 
 | Parameter | Description | Default |
 |-----------|-------------|---------|
 | `--model` | Whisper model size. | `small` |
-| `--language` | Source language code or `auto` | `en` |
+| `--language` | Source language code or `auto` | `auto` |
-| `--task` | `transcribe` or `translate` | `transcribe` |
+| `--task` | Set to `translate` to translate to english | `transcribe` |
+| `--target-language` | [BETA] Translation language target. Ex: `fr` | `None` |
 | `--backend` | Processing backend | `simulstreaming` |
 | `--min-chunk-size` | Minimum audio chunk size (seconds) | `1.0` |
 | `--no-vac` | Disable Voice Activity Controller | `False` |
@@ -159,16 +164,12 @@ The package includes an HTML/JavaScript implementation [here](https://github.com
 | `--port` | Server port | `8000` |
 | `--ssl-certfile` | Path to the SSL certificate file (for HTTPS support) | `None` |
 | `--ssl-keyfile` | Path to the SSL private key file (for HTTPS support) | `None` |
-
+| `--pcm-input` | raw PCM (s16le) data is expected as input and FFmpeg will be bypassed. | `False` |
-
-| WhisperStreaming backend options | Description | Default |
-|-----------|-------------|---------|
-| `--confidence-validation` | Use confidence scores for faster validation | `False` |
-| `--buffer_trimming` | Buffer trimming strategy (`sentence` or `segment`) | `segment` |
 
 
 | SimulStreaming backend options | Description | Default |
 |-----------|-------------|---------|
+| `--disable-fast-encoder` | Disable Faster Whisper or MLX Whisper backends for the encoder (if installed). Inference can be slower but helpful when GPU memory is limited | `False` |
 | `--frame-threshold` | AlignAtt frame threshold (lower = faster, higher = more accurate) | `25` |
 | `--beams` | Number of beams for beam search (1 = greedy decoding) | `1` |
 | `--decoder` | Force decoder type (`beam` or `greedy`) | `auto` |
@@ -180,14 +181,28 @@ The package includes an HTML/JavaScript implementation [here](https://github.com
 | `--static-init-prompt` | Static prompt that doesn't scroll | `None` |
 | `--max-context-tokens` | Maximum context tokens | `None` |
 | `--model-path` | Direct path to .pt model file. Download it if not found | `./base.pt` |
-| `--preloaded-model-count` | Optional. Number of models to preload in memory to speed up loading (set up to the expected number of concurrent users) | `1` |
+| `--preload-model-count` | Optional. Number of models to preload in memory to speed up loading (set up to the expected number of concurrent users) | `1` |
+
+
+| WhisperStreaming backend options | Description | Default |
+|-----------|-------------|---------|
+| `--confidence-validation` | Use confidence scores for faster validation | `False` |
+| `--buffer_trimming` | Buffer trimming strategy (`sentence` or `segment`) | `segment` |
 
 | Diarization options | Description | Default |
 |-----------|-------------|---------|
 | `--diarization` | Enable speaker identification | `False` |
-| `--punctuation-split` | Use punctuation to improve speaker boundaries | `True` |
+| `--diarization-backend` |  `diart` or `sortformer` | `sortformer` |
-| `--segmentation-model` | Hugging Face model ID for pyannote.audio segmentation model. [Available models](https://github.com/juanmc2005/diart/tree/main?tab=readme-ov-file#pre-trained-models) | `pyannote/segmentation-3.0` |
+| `--disable-punctuation-split` |  Disable punctuation based splits. See #214 | `False` |
-| `--embedding-model` | Hugging Face model ID for pyannote.audio embedding model. [Available models](https://github.com/juanmc2005/diart/tree/main?tab=readme-ov-file#pre-trained-models) | `speechbrain/spkrec-ecapa-voxceleb` |
+| `--segmentation-model` | Hugging Face model ID for Diart segmentation model. [Available models](https://github.com/juanmc2005/diart/tree/main?tab=readme-ov-file#pre-trained-models) | `pyannote/segmentation-3.0` |
+| `--embedding-model` | Hugging Face model ID for Diart embedding model. [Available models](https://github.com/juanmc2005/diart/tree/main?tab=readme-ov-file#pre-trained-models) | `speechbrain/spkrec-ecapa-voxceleb` |
+
+
+> For diarization using Diart, you need access to pyannote.audio models:
+> 1. [Accept user conditions](https://huggingface.co/pyannote/segmentation) for the `pyannote/segmentation` model
+> 2. [Accept user conditions](https://huggingface.co/pyannote/segmentation-3.0) for the `pyannote/segmentation-3.0` model
+> 3. [Accept user conditions](https://huggingface.co/pyannote/embedding) for the `pyannote/embedding` model
+>4. Login with HuggingFace: `huggingface-cli login`
 
 ### 🚀 Deployment Guide
 
@@ -216,19 +231,39 @@ To deploy WhisperLiveKit in production:
 
 4. **HTTPS Support**: For secure deployments, use "wss://" instead of "ws://" in WebSocket URL
 
-### 🐋 Docker
+## 🐋 Docker
 
-A Dockerfile is provided which allows re-use of Python package installation options. Create a reusable image with only the basics and then run as a named container:
+Deploy the application easily using Docker with GPU or CPU support.
 
+### Prerequisites
+- Docker installed on your system
+- For GPU support: NVIDIA Docker runtime installed
+
+### Quick Start
+
+**With GPU acceleration (recommended):**
 ```bash
-docker build -t whisperlivekit-defaults .
+docker build -t wlk .
-docker create --gpus all --name whisperlivekit -p 8000:8000 whisperlivekit-defaults --model base
+docker run --gpus all -p 8000:8000 --name wlk wlk
-docker start -i whisperlivekit
 ```
 
-> **Note**: For **large** models, ensure that your **docker runtime** has enough **memory** available
+**CPU only:**
+```bash
+docker build -f Dockerfile.cpu -t wlk .
+docker run -p 8000:8000 --name wlk wlk
+```
+
+### Advanced Usage
+
+**Custom configuration:**
+```bash
+# Example with custom model and language
+docker run --gpus all -p 8000:8000 --name wlk wlk --model large-v3 --language fr
+```
+
+### Memory Requirements
+- **Large models**: Ensure your Docker runtime has sufficient memory allocated
 
-> **Note**: If you're running on a system without NVIDIA GPU support (such as Mac with Apple Silicon or any system without CUDA capabilities), you need to **remove the `--gpus all` flag** from the `docker create` command. Without GPU acceleration, transcription will use CPU only, which may be significantly slower. Consider using small models for better performance on CPU-only systems.
 
 #### Customization
 

ReadmeJP.md

@@ -0,0 +1,258 @@
+<h1 align="center">WhisperLiveKit</h1>
+
+<p align="center">
+<img src="https://raw.githubusercontent.com/QuentinFuxa/WhisperLiveKit/refs/heads/main/demo.png" alt="WhisperLiveKit Demo" width="730">
+</p>
+
+<p align="center"><b>話者識別機能付き、リアルタイム、完全ローカルな音声テキスト変換</b></p>
+
+<p align="center">
+<a href="https://pypi.org/project/whisperlivekit/"><img alt="PyPI Version" src="https://img.shields.io/pypi/v/whisperlivekit?color=g"></a>
+<a href="https://pepy.tech/project/whisperlivekit"><img alt="PyPI Downloads" src="https://static.pepy.tech/personalized-badge/whisperlivekit?period=total&units=international_system&left_color=grey&right_color=brightgreen&left_text=installations"></a>
+<a href="https://pypi.org/project/whisperlivekit/"><img alt="Python Versions" src="https://img.shields.io/badge/python-3.9--3.13-dark_green"></a>
+<a href="https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/LICENSE"><img alt="License" src="https://img.shields.io/badge/License-MIT/Dual Licensed-dark_green"></a>
+</p>
+
+すぐに使えるバックエンド+サーバーとシンプルなフロントエンドで、リアルタイムの音声文字起こしをブラウザに直接提供します。✨
+
+#### 主要な研究による技術：
+
+- [SimulStreaming](https://github.com/ufal/SimulStreaming) (SOTA 2025) - AlignAttポリシーによる超低遅延文字起こし
+- [WhisperStreaming](https://github.com/ufal/whisper_streaming) (SOTA 2023) - LocalAgreementポリシーによる低遅延文字起こし
+- [Streaming Sortformer](https://arxiv.org/abs/2507.18446) (SOTA 2025) - 高度なリアルタイム話者ダイアライゼーション
+- [Diart](https://github.com/juanmc2005/diart) (SOTA 2021) - リアルタイム話者ダイアライゼーション
+- [Silero VAD](https://github.com/snakers4/silero-vad) (2024) - エンタープライズグレードの音声区間検出
+
+> **なぜ各音声バッチで単純なWhisperモデルを実行しないのか？** Whisperは完全な発話向けに設計されており、リアルタイムのチャンク向けではありません。小さなセグメントを処理するとコンテキストが失われ、単語が音節の途中で途切れ、質の悪い文字起こしになります。WhisperLiveKitは、インテリジェントなバッファリングとインクリメンタルな処理のために、最先端の同時音声研究を利用しています。
+
+### アーキテクチャ
+
+<img alt="Architecture" src="https://raw.githubusercontent.com/QuentinFuxa/WhisperLiveKit/refs/heads/main/architecture.png" />
+
+*バックエンドは複数の同時ユーザーをサポートします。音声が検出されない場合、音声区間検出がオーバーヘッドを削減します。*
+
+### インストールとクイックスタート
+
+```bash
+pip install whisperlivekit
+```
+
+>  **FFmpegが必要です** WhisperLiveKitを使用する前にインストールする必要があります。
+>
+> | OS | インストール方法 |
+> |-----------|-------------|
+>  | Ubuntu/Debian | `sudo apt install ffmpeg` |
+> | MacOS | `brew install ffmpeg` |
+> | Windows | https://ffmpeg.org/download.html から.exeをダウンロードし、PATHに追加 |
+
+#### クイックスタート
+1. **文字起こしサーバーを起動します:**
+   ```bash
+   whisperlivekit-server --model base --language en
+   ```
+
+2. **ブラウザを開き** `http://localhost:8000` にアクセスします。話し始めると、あなたの言葉がリアルタイムで表示されます！
+
+
+> - 利用可能なすべての言語のリストについては、[tokenizer.py](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/whisperlivekit/simul_whisper/whisper/tokenizer.py) を参照してください。
+> - HTTPSの要件については、**パラメータ**セクションのSSL設定オプションを参照してください。
+
+#### オプションの依存関係
+
+| オプション | `pip install` |
+|-----------|-------------|
+| **Sortformerによる話者ダイアライゼーション** | `git+https://github.com/NVIDIA/NeMo.git@main#egg=nemo_toolkit[asr]` |
+| Diartによる話者ダイアライゼーション | `diart` |
+| オリジナルのWhisperバックエンド | `whisper` |
+| タイムスタンプ改善バックエンド | `whisper-timestamped` |
+| Apple Silicon最適化バックエンド | `mlx-whisper` |
+| OpenAI APIバックエンド | `openai` |
+
+それらの使用方法については、以下の**パラメータと設定**を参照してください。
+
+### 使用例
+
+**コマンドラインインターフェース**: 様々なオプションで文字起こしサーバーを起動します:
+
+```bash
+# デフォルト(small)より良いモデルを使用
+whisperlivekit-server --model large-v3
+
+# ダイアライゼーションと言語を指定した高度な設定
+whisperlivekit-server --host 0.0.0.0 --port 8000 --model medium --diarization --language fr
+```
+
+**Python API連携**: 関数やクラスの使用方法のより完全な例については、[basic_server](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/whisperlivekit/basic_server.py) を確認してください。
+
+```python
+from whisperlivekit import TranscriptionEngine, AudioProcessor, parse_args
+from fastapi import FastAPI, WebSocket, WebSocketDisconnect
+from fastapi.responses import HTMLResponse
+from contextlib import asynccontextmanager
+import asyncio
+
+transcription_engine = None
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    global transcription_engine
+    transcription_engine = TranscriptionEngine(model="medium", diarization=True, lan="en")
+    yield
+
+app = FastAPI(lifespan=lifespan)
+
+async def handle_websocket_results(websocket: WebSocket, results_generator):
+    async for response in results_generator:
+        await websocket.send_json(response)
+    await websocket.send_json({"type": "ready_to_stop"})
+
+@app.websocket("/asr")
+async def websocket_endpoint(websocket: WebSocket):
+    global transcription_engine
+
+    # 接続ごとに新しいAudioProcessorを作成し、共有エンジンを渡す
+    audio_processor = AudioProcessor(transcription_engine=transcription_engine)
+    results_generator = await audio_processor.create_tasks()
+    results_task = asyncio.create_task(handle_websocket_results(websocket, results_generator))
+    await websocket.accept()
+    while True:
+        message = await websocket.receive_bytes()
+        await audio_processor.process_audio(message)
+```
+
+**フロントエンド実装**: パッケージにはHTML/JavaScript実装が[ここ](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/whisperlivekit/web/live_transcription.html)に含まれています。`from whisperlivekit import get_web_interface_html` & `page = get_web_interface_html()` を使ってインポートすることもできます。
+
+
+## パラメータと設定
+
+重要なパラメータのリストを変更できます。しかし、何を*変更すべき*でしょうか？
+- `--model` サイズ。リストと推奨事項は[こちら](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/available_models.md)
+- `--language`。リストは[こちら](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/whisperlivekit/simul_whisper/whisper/tokenizer.py)。`auto`を使用すると、モデルは自動的に言語を検出しようとしますが、英語に偏る傾向があります。
+- `--backend`？ `simulstreaming`が正しく動作しない場合や、デュアルライセンス要件を避けたい場合は`--backend faster-whisper`に切り替えることができます。
+- `--warmup-file`、もしあれば
+- `--host`, `--port`, `--ssl-certfile`, `--ssl-keyfile`、サーバーをセットアップする場合
+- `--diarization`、使用したい場合。
+
+残りは推奨しません。しかし、以下があなたのオプションです。
+
+| パラメータ | 説明 | デフォルト |
+|-----------|-------------|---------|
+| `--model` | Whisperモデルのサイズ。 | `small` |
+| `--language` | ソース言語コードまたは`auto` | `auto` |
+| `--task` | `transcribe`または`translate` | `transcribe` |
+| `--backend` | 処理バックエンド | `simulstreaming` |
+| `--min-chunk-size` | 最小音声チャンクサイズ（秒） | `1.0` |
+| `--no-vac` | 音声アクティビティコントローラーを無効化 | `False` |
+| `--no-vad` | 音声区間検出を無効化 | `False` |
+| `--warmup-file` | モデルのウォームアップ用音声ファイルパス | `jfk.wav` |
+| `--host` | サーバーホストアドレス | `localhost` |
+| `--port` | サーバーポート | `8000` |
+| `--ssl-certfile` | SSL証明書ファイルへのパス（HTTPSサポート用） | `None` |
+| `--ssl-keyfile` | SSL秘密鍵ファイルへのパス（HTTPSサポート用） | `None` |
+
+
+| WhisperStreamingバックエンドオプション | 説明 | デフォルト |
+|-----------|-------------|---------|
+| `--confidence-validation` | 高速な検証のために信頼スコアを使用 | `False` |
+| `--buffer_trimming` | バッファトリミング戦略（`sentence`または`segment`） | `segment` |
+
+
+| SimulStreamingバックエンドオプション | 説明 | デフォルト |
+|-----------|-------------|---------|
+| `--frame-threshold` | AlignAttフレームしきい値（低いほど速く、高いほど正確） | `25` |
+| `--beams` | ビームサーチのビーム数（1 = 貪欲デコーディング） | `1` |
+| `--decoder` | デコーダタイプを強制（`beam`または`greedy`） | `auto` |
+| `--audio-max-len` | 最大音声バッファ長（秒） | `30.0` |
+| `--audio-min-len` | 処理する最小音声長（秒） | `0.0` |
+| `--cif-ckpt-path` | 単語境界検出用CIFモデルへのパス | `None` |
+| `--never-fire` | 未完了の単語を決して切り捨てない | `False` |
+| `--init-prompt` | モデルの初期プロンプト | `None` |
+| `--static-init-prompt` | スクロールしない静的プロンプト | `None` |
+| `--max-context-tokens` | 最大コンテキストトークン数 | `None` |
+| `--model-path` | .ptモデルファイルへの直接パス。見つからない場合はダウンロード | `./base.pt` |
+| `--preloaded-model-count` | オプション。メモリにプリロードするモデルの数（予想される同時ユーザー数まで設定） | `1` |
+
+| ダイアライゼーションオプション | 説明 | デフォルト |
+|-----------|-------------|---------|
+| `--diarization` | 話者識別を有効化 | `False` |
+| `--diarization-backend` | `diart`または`sortformer` | `sortformer` |
+| `--segmentation-model` | DiartセグメンテーションモデルのHugging FaceモデルID。[利用可能なモデル](https://github.com/juanmc2005/diart/tree/main?tab=readme-ov-file#pre-trained-models) | `pyannote/segmentation-3.0` |
+| `--embedding-model` | Diart埋め込みモデルのHugging FaceモデルID。[利用可能なモデル](https://github.com/juanmc2005/diart/tree/main?tab=readme-ov-file#pre-trained-models) | `speechbrain/spkrec-ecapa-voxceleb` |
+
+
+> Diartを使用したダイアライゼーションには、pyannote.audioモデルへのアクセスが必要です：
+> 1. `pyannote/segmentation`モデルの[ユーザー条件に同意](https://huggingface.co/pyannote/segmentation)
+> 2. `pyannote/segmentation-3.0`モデルの[ユーザー条件に同意](https://huggingface.co/pyannote/segmentation-3.0)
+> 3. `pyannote/embedding`モデルの[ユーザー条件に同意](https://huggingface.co/pyannote/embedding)
+>4. HuggingFaceでログイン: `huggingface-cli login`
+
+### 🚀 デプロイガイド
+
+WhisperLiveKitを本番環境にデプロイするには：
+
+1. **サーバーセットアップ**: 本番用ASGIサーバーをインストールし、複数のワーカーで起動します
+   ```bash
+   pip install uvicorn gunicorn
+   gunicorn -k uvicorn.workers.UvicornWorker -w 4 your_app:app
+   ```
+
+2. **フロントエンド**: カスタマイズした`html`のバージョンをホストし、WebSocket接続が正しくポイントするようにします
+
+3. **Nginx設定** (本番環境で推奨):
+    ```nginx
+   server {
+       listen 80;
+       server_name your-domain.com;
+        location / {
+            proxy_pass http://localhost:8000;
+            proxy_set_header Upgrade $http_upgrade;
+            proxy_set_header Connection "upgrade";
+            proxy_set_header Host $host;
+    }}
+    ```
+
+4. **HTTPSサポート**: 安全なデプロイメントのために、WebSocket URLで "ws://" の代わりに "wss://" を使用します
+
+## 🐋 Docker
+
+GPUまたはCPUサポート付きでDockerを使用してアプリケーションを簡単にデプロイします。
+
+### 前提条件
+- Dockerがシステムにインストールされていること
+- GPUサポートの場合: NVIDIA Dockerランタイムがインストールされていること
+
+### クイックスタート
+
+**GPUアクセラレーション付き (推奨):**
+```bash
+docker build -t wlk .
+docker run --gpus all -p 8000:8000 --name wlk wlk
+```
+
+**CPUのみ:**
+```bash
+docker build -f Dockerfile.cpu -t wlk .
+docker run -p 8000:8000 --name wlk wlk
+```
+
+### 高度な使用法
+
+**カスタム設定:**
+```bash
+# カスタムモデルと言語の例
+docker run --gpus all -p 8000:8000 --name wlk wlk --model large-v3 --language fr
+```
+
+### メモリ要件
+- **大規模モデル**: Dockerランタイムに十分なメモリが割り当てられていることを確認してください
+
+
+#### カスタマイズ
+
+- `--build-arg` オプション:
+  - `EXTRAS="whisper-timestamped"` - イメージのインストールにエクストラを追加します（スペースなし）。必要なコンテナオプションを設定することを忘れないでください！
+  - `HF_PRECACHE_DIR="./.cache/"` - 初回起動を高速化するためにモデルキャッシュをプリロードします
+  - `HF_TKN_FILE="./token"` - ゲート付きモデルをダウンロードするためにHugging Face Hubアクセストークンを追加します
+
+## 🔮 ユースケース
+会議の文字起こしのためにリアルタイムで議論をキャプチャする、聴覚障害のあるユーザーがアクセシビリティツールを通じて会話を追うのを助ける、コンテンツ作成のためにポッドキャストやビデオを自動的に文字起こしする、カスタマーサービスのために話者識別付きでサポートコールを文字起こしする...

available_models.md

@@ -0,0 +1,73 @@
+# Available model sizes:
+
+- tiny.en (english only)
+- tiny
+- base.en (english only)
+- base
+- small.en (english only)
+- small
+- medium.en (english only)
+- medium
+- large-v1
+- large-v2
+- large-v3
+- large-v3-turbo
+
+## How to choose?
+
+### Language Support
+- **English only**: Use `.en` models for better accuracy and faster processing when you only need English transcription
+- **Multilingual**: Do not use `.en` models.
+
+### Resource Constraints
+- **Limited GPU/CPU or need for very low latency**: Choose `small` or smaller models
+  - `tiny`: Fastest, lowest resource usage, acceptable quality for simple audio
+  - `base`: Good balance of speed and accuracy for basic use cases
+  - `small`: Better accuracy while still being resource-efficient
+- **Good resources available**: Use `large` models for best accuracy
+  - `large-v2`: Excellent accuracy, good multilingual support
+  - `large-v3`: Best overall accuracy and language support
+
+### Special Cases
+- **No translation needed**: Use `large-v3-turbo`
+  - Same transcription quality as `large-v2` but significantly faster
+  - **Important**: Does not translate correctly, only transcribes
+
+### Model Comparison Table
+
+| Model | Speed | Accuracy | Multilingual | Translation | Best Use Case |
+|-------|--------|----------|--------------|-------------|---------------|
+| tiny(.en) | Fastest | Basic | Yes/No | Yes/No | Real-time, low resources |
+| base(.en) | Fast | Good | Yes/No | Yes/No | Balanced performance |
+| small(.en) | Medium | Better | Yes/No | Yes/No | Quality on limited hardware |
+| medium(.en) | Slow | High | Yes/No | Yes/No | High quality, moderate resources |
+| large-v2 | Slowest | Excellent | Yes | Yes | Best overall quality |
+| large-v3 | Slowest | Excellent | Yes | Yes | Maximum accuracy |
+| large-v3-turbo | Fast | Excellent | Yes | No | Fast, high-quality transcription |
+
+### Additional Considerations
+
+**Model Performance**:
+- Accuracy improves significantly from tiny to large models
+- English-only models are ~10-15% more accurate for English audio
+- Newer versions (v2, v3) have better punctuation and formatting
+
+**Hardware Requirements**:
+- `tiny`: ~1GB VRAM
+- `base`: ~1GB VRAM  
+- `small`: ~2GB VRAM
+- `medium`: ~5GB VRAM
+- `large`: ~10GB VRAM
+- `large‑v3‑turbo`: ~6GB VRAM
+
+**Audio Quality Impact**:
+- Clean, clear audio: smaller models may suffice
+- Noisy, accented, or technical audio: larger models recommended
+- Phone/low-quality audio: use at least `small` model
+
+### Quick Decision Tree
+1. English only? → Add `.en` to your choice
+2. Limited resources or need speed? → `small` or smaller
+3. Good hardware and want best quality? → `large-v3`
+4. Need fast, high-quality transcription without translation? → `large-v3-turbo`
+5. Need translation capabilities? → `large-v2` or `large-v3` (avoid turbo)

chrome-extension/README.md

@@ -0,0 +1,17 @@
+## WhisperLiveKit Chrome Extension v0.1.0
+Capture the audio of your current tab, transcribe or translate it using WhisperliveKit. **Still unstable**
+
+<img src="https://raw.githubusercontent.com/QuentinFuxa/WhisperLiveKit/refs/heads/main/chrome-extension/demo-extension.png" alt="WhisperLiveKit Demo" width="730">
+
+## Running this extension
+1. Clone this repository.
+2. Load this directory in Chrome as an unpacked extension.
+
+
+## Devs:
+- Impossible to capture audio from tabs if extension is a pannel, unfortunately: 
+- https://issues.chromium.org/issues/40926394
+- https://groups.google.com/a/chromium.org/g/chromium-extensions/c/DET2SXCFnDg
+- https://issues.chromium.org/issues/40916430
+
+- To capture microphone in an extension, there are tricks: https://github.com/justinmann/sidepanel-audio-issue , https://medium.com/@lynchee.owo/how-to-enable-microphone-access-in-chrome-extensions-by-code-924295170080 (comments)

chrome-extension/background.js

@@ -0,0 +1,9 @@
+chrome.runtime.onInstalled.addListener((details) => {
+    if (details.reason.search(/install/g) === -1) {
+        return
+    }
+    chrome.tabs.create({
+        url: chrome.runtime.getURL("welcome.html"),
+        active: true
+    })
+})

chrome-extension/live_transcription.js

@@ -0,0 +1,669 @@
+/* Theme, WebSocket, recording, rendering logic extracted from inline script and adapted for segmented theme control and WS caption */
+let isRecording = false;
+let websocket = null;
+let recorder = null;
+let chunkDuration = 100;
+let websocketUrl = "ws://localhost:8000/asr";
+let userClosing = false;
+let wakeLock = null;
+let startTime = null;
+let timerInterval = null;
+let audioContext = null;
+let analyser = null;
+let microphone = null;
+let waveCanvas = document.getElementById("waveCanvas");
+let waveCtx = waveCanvas.getContext("2d");
+let animationFrame = null;
+let waitingForStop = false;
+let lastReceivedData = null;
+let lastSignature = null;
+let availableMicrophones = [];
+let selectedMicrophoneId = null;
+
+waveCanvas.width = 60 * (window.devicePixelRatio || 1);
+waveCanvas.height = 30 * (window.devicePixelRatio || 1);
+waveCtx.scale(window.devicePixelRatio || 1, window.devicePixelRatio || 1);
+
+const statusText = document.getElementById("status");
+const recordButton = document.getElementById("recordButton");
+const chunkSelector = document.getElementById("chunkSelector");
+const websocketInput = document.getElementById("websocketInput");
+const websocketDefaultSpan = document.getElementById("wsDefaultUrl");
+const linesTranscriptDiv = document.getElementById("linesTranscript");
+const timerElement = document.querySelector(".timer");
+const themeRadios = document.querySelectorAll('input[name="theme"]');
+const microphoneSelect = document.getElementById("microphoneSelect");
+const settingsToggle = document.getElementById("settingsToggle");
+const settingsDiv = document.querySelector(".settings");
+
+
+
+chrome.runtime.onInstalled.addListener((details) => {
+    if (details.reason.search(/install/g) === -1) {
+        return
+    }
+    chrome.tabs.create({
+        url: chrome.runtime.getURL("welcome.html"),
+        active: true
+    })
+})
+
+function getWaveStroke() {
+  const styles = getComputedStyle(document.documentElement);
+  const v = styles.getPropertyValue("--wave-stroke").trim();
+  return v || "#000";
+}
+
+let waveStroke = getWaveStroke();
+function updateWaveStroke() {
+  waveStroke = getWaveStroke();
+}
+
+function applyTheme(pref) {
+  if (pref === "light") {
+    document.documentElement.setAttribute("data-theme", "light");
+  } else if (pref === "dark") {
+    document.documentElement.setAttribute("data-theme", "dark");
+  } else {
+    document.documentElement.removeAttribute("data-theme");
+  }
+  updateWaveStroke();
+}
+
+// Persisted theme preference
+const savedThemePref = localStorage.getItem("themePreference") || "system";
+applyTheme(savedThemePref);
+if (themeRadios.length) {
+  themeRadios.forEach((r) => {
+    r.checked = r.value === savedThemePref;
+    r.addEventListener("change", () => {
+      if (r.checked) {
+        localStorage.setItem("themePreference", r.value);
+        applyTheme(r.value);
+      }
+    });
+  });
+}
+
+// React to OS theme changes when in "system" mode
+const darkMq = window.matchMedia && window.matchMedia("(prefers-color-scheme: dark)");
+const handleOsThemeChange = () => {
+  const pref = localStorage.getItem("themePreference") || "system";
+  if (pref === "system") updateWaveStroke();
+};
+if (darkMq && darkMq.addEventListener) {
+  darkMq.addEventListener("change", handleOsThemeChange);
+} else if (darkMq && darkMq.addListener) {
+  // deprecated, but included for Safari compatibility
+  darkMq.addListener(handleOsThemeChange);
+}
+
+async function enumerateMicrophones() {
+  try {
+      const micPermission = await navigator.permissions.query({
+    name: "microphone",
+  });
+  
+    const stream = await navigator.mediaDevices.getUserMedia({ audio: true });
+    stream.getTracks().forEach(track => track.stop());
+
+    const devices = await navigator.mediaDevices.enumerateDevices();
+    availableMicrophones = devices.filter(device => device.kind === 'audioinput');
+
+    populateMicrophoneSelect();
+    console.log(`Found ${availableMicrophones.length} microphone(s)`);
+  } catch (error) {
+    console.error('Error enumerating microphones:', error);
+    statusText.textContent = "Error accessing microphones. Please grant permission.";
+  }
+}
+
+function populateMicrophoneSelect() {
+  if (!microphoneSelect) return;
+
+  microphoneSelect.innerHTML = '<option value="">Default Microphone</option>';
+
+  availableMicrophones.forEach((device, index) => {
+    const option = document.createElement('option');
+    option.value = device.deviceId;
+    option.textContent = device.label || `Microphone ${index + 1}`;
+    microphoneSelect.appendChild(option);
+  });
+
+  const savedMicId = localStorage.getItem('selectedMicrophone');
+  if (savedMicId && availableMicrophones.some(mic => mic.deviceId === savedMicId)) {
+    microphoneSelect.value = savedMicId;
+    selectedMicrophoneId = savedMicId;
+  }
+}
+
+function handleMicrophoneChange() {
+  selectedMicrophoneId = microphoneSelect.value || null;
+  localStorage.setItem('selectedMicrophone', selectedMicrophoneId || '');
+
+  const selectedDevice = availableMicrophones.find(mic => mic.deviceId === selectedMicrophoneId);
+  const deviceName = selectedDevice ? selectedDevice.label : 'Default Microphone';
+
+  console.log(`Selected microphone: ${deviceName}`);
+  statusText.textContent = `Microphone changed to: ${deviceName}`;
+
+  if (isRecording) {
+    statusText.textContent = "Switching microphone... Please wait.";
+    stopRecording().then(() => {
+      setTimeout(() => {
+        toggleRecording();
+      }, 1000);
+    });
+  }
+}
+
+// Helpers
+function fmt1(x) {
+  const n = Number(x);
+  return Number.isFinite(n) ? n.toFixed(1) : x;
+}
+
+// Default WebSocket URL computation
+const host = window.location.hostname || "localhost";
+const port = window.location.port;
+const protocol = window.location.protocol === "https:" ? "wss" : "ws";
+const defaultWebSocketUrl = websocketUrl;
+
+// Populate default caption and input
+if (websocketDefaultSpan) websocketDefaultSpan.textContent = defaultWebSocketUrl;
+websocketInput.value = defaultWebSocketUrl;
+websocketUrl = defaultWebSocketUrl;
+
+// Optional chunk selector (guard for presence)
+if (chunkSelector) {
+  chunkSelector.addEventListener("change", () => {
+    chunkDuration = parseInt(chunkSelector.value);
+  });
+}
+
+// WebSocket input change handling
+websocketInput.addEventListener("change", () => {
+  const urlValue = websocketInput.value.trim();
+  if (!urlValue.startsWith("ws://") && !urlValue.startsWith("wss://")) {
+    statusText.textContent = "Invalid WebSocket URL (must start with ws:// or wss://)";
+    return;
+  }
+  websocketUrl = urlValue;
+  statusText.textContent = "WebSocket URL updated. Ready to connect.";
+});
+
+function setupWebSocket() {
+  return new Promise((resolve, reject) => {
+    try {
+      websocket = new WebSocket(websocketUrl);
+    } catch (error) {
+      statusText.textContent = "Invalid WebSocket URL. Please check and try again.";
+      reject(error);
+      return;
+    }
+
+    websocket.onopen = () => {
+      statusText.textContent = "Connected to server.";
+      resolve();
+    };
+
+    websocket.onclose = () => {
+      if (userClosing) {
+        if (waitingForStop) {
+          statusText.textContent = "Processing finalized or connection closed.";
+          if (lastReceivedData) {
+            renderLinesWithBuffer(
+              lastReceivedData.lines || [],
+              lastReceivedData.buffer_diarization || "",
+              lastReceivedData.buffer_transcription || "",
+              0,
+              0,
+              true
+            );
+          }
+        }
+      } else {
+        statusText.textContent = "Disconnected from the WebSocket server. (Check logs if model is loading.)";
+        if (isRecording) {
+          stopRecording();
+        }
+      }
+      isRecording = false;
+      waitingForStop = false;
+      userClosing = false;
+      lastReceivedData = null;
+      websocket = null;
+      updateUI();
+    };
+
+    websocket.onerror = () => {
+      statusText.textContent = "Error connecting to WebSocket.";
+      reject(new Error("Error connecting to WebSocket"));
+    };
+
+    websocket.onmessage = (event) => {
+      const data = JSON.parse(event.data);
+
+      if (data.type === "ready_to_stop") {
+        console.log("Ready to stop received, finalizing display and closing WebSocket.");
+        waitingForStop = false;
+
+        if (lastReceivedData) {
+          renderLinesWithBuffer(
+            lastReceivedData.lines || [],
+            lastReceivedData.buffer_diarization || "",
+            lastReceivedData.buffer_transcription || "",
+            0,
+            0,
+            true
+          );
+        }
+        statusText.textContent = "Finished processing audio! Ready to record again.";
+        recordButton.disabled = false;
+
+        if (websocket) {
+          websocket.close();
+        }
+        return;
+      }
+
+      lastReceivedData = data;
+
+      const {
+        lines = [],
+        buffer_transcription = "",
+        buffer_diarization = "",
+        remaining_time_transcription = 0,
+        remaining_time_diarization = 0,
+        status = "active_transcription",
+      } = data;
+
+      renderLinesWithBuffer(
+        lines,
+        buffer_diarization,
+        buffer_transcription,
+        remaining_time_diarization,
+        remaining_time_transcription,
+        false,
+        status
+      );
+    };
+  });
+}
+
+function renderLinesWithBuffer(
+  lines,
+  buffer_diarization,
+  buffer_transcription,
+  remaining_time_diarization,
+  remaining_time_transcription,
+  isFinalizing = false,
+  current_status = "active_transcription"
+) {
+  if (current_status === "no_audio_detected") {
+    linesTranscriptDiv.innerHTML =
+      "<p style='text-align: center; color: var(--muted); margin-top: 20px;'><em>No audio detected...</em></p>";
+    return;
+  }
+
+  const showLoading = !isFinalizing && (lines || []).some((it) => it.speaker == 0);
+  const showTransLag = !isFinalizing && remaining_time_transcription > 0;
+  const showDiaLag = !isFinalizing && !!buffer_diarization && remaining_time_diarization > 0;
+  const signature = JSON.stringify({
+    lines: (lines || []).map((it) => ({ speaker: it.speaker, text: it.text, start: it.start, end: it.end })),
+    buffer_transcription: buffer_transcription || "",
+    buffer_diarization: buffer_diarization || "",
+    status: current_status,
+    showLoading,
+    showTransLag,
+    showDiaLag,
+    isFinalizing: !!isFinalizing,
+  });
+  if (lastSignature === signature) {
+    const t = document.querySelector(".lag-transcription-value");
+    if (t) t.textContent = fmt1(remaining_time_transcription);
+    const d = document.querySelector(".lag-diarization-value");
+    if (d) d.textContent = fmt1(remaining_time_diarization);
+    const ld = document.querySelector(".loading-diarization-value");
+    if (ld) ld.textContent = fmt1(remaining_time_diarization);
+    return;
+  }
+  lastSignature = signature;
+
+  const linesHtml = (lines || [])
+    .map((item, idx) => {
+      let timeInfo = "";
+      if (item.start !== undefined && item.end !== undefined) {
+        timeInfo = ` ${item.start} - ${item.end}`;
+      }
+
+      let speakerLabel = "";
+      if (item.speaker === -2) {
+        speakerLabel = `<span class="silence">Silence<span id='timeInfo'>${timeInfo}</span></span>`;
+      } else if (item.speaker == 0 && !isFinalizing) {
+        speakerLabel = `<span class='loading'><span class="spinner"></span><span id='timeInfo'><span class="loading-diarization-value">${fmt1(
+          remaining_time_diarization
+        )}</span> second(s) of audio are undergoing diarization</span></span>`;
+      } else if (item.speaker !== 0) {
+        speakerLabel = `<span id="speaker">Speaker ${item.speaker}<span id='timeInfo'>${timeInfo}</span></span>`;
+      }
+
+      let currentLineText = item.text || "";
+
+      if (idx === lines.length - 1) {
+        if (!isFinalizing && item.speaker !== -2) {
+          if (remaining_time_transcription > 0) {
+            speakerLabel += `<span class="label_transcription"><span class="spinner"></span>Lag <span id='timeInfo'><span class="lag-transcription-value">${fmt1(
+              remaining_time_transcription
+            )}</span>s</span></span>`;
+          }
+          if (buffer_diarization && remaining_time_diarization > 0) {
+            speakerLabel += `<span class="label_diarization"><span class="spinner"></span>Lag<span id='timeInfo'><span class="lag-diarization-value">${fmt1(
+              remaining_time_diarization
+            )}</span>s</span></span>`;
+          }
+        }
+
+        if (buffer_diarization) {
+          if (isFinalizing) {
+            currentLineText +=
+              (currentLineText.length > 0 && buffer_diarization.trim().length > 0 ? " " : "") + buffer_diarization.trim();
+          } else {
+            currentLineText += `<span class="buffer_diarization">${buffer_diarization}</span>`;
+          }
+        }
+        if (buffer_transcription) {
+          if (isFinalizing) {
+            currentLineText +=
+              (currentLineText.length > 0 && buffer_transcription.trim().length > 0 ? " " : "") +
+              buffer_transcription.trim();
+          } else {
+            currentLineText += `<span class="buffer_transcription">${buffer_transcription}</span>`;
+          }
+        }
+      }
+
+      return currentLineText.trim().length > 0 || speakerLabel.length > 0
+        ? `<p>${speakerLabel}<br/><div class='textcontent'>${currentLineText}</div></p>`
+        : `<p>${speakerLabel}<br/></p>`;
+    })
+    .join("");
+
+  linesTranscriptDiv.innerHTML = linesHtml;
+  window.scrollTo({ top: document.body.scrollHeight, behavior: "smooth" });
+}
+
+function updateTimer() {
+  if (!startTime) return;
+
+  const elapsed = Math.floor((Date.now() - startTime) / 1000);
+  const minutes = Math.floor(elapsed / 60).toString().padStart(2, "0");
+  const seconds = (elapsed % 60).toString().padStart(2, "0");
+  timerElement.textContent = `${minutes}:${seconds}`;
+}
+
+function drawWaveform() {
+  if (!analyser) return;
+
+  const bufferLength = analyser.frequencyBinCount;
+  const dataArray = new Uint8Array(bufferLength);
+  analyser.getByteTimeDomainData(dataArray);
+
+  waveCtx.clearRect(
+    0,
+    0,
+    waveCanvas.width / (window.devicePixelRatio || 1),
+    waveCanvas.height / (window.devicePixelRatio || 1)
+  );
+  waveCtx.lineWidth = 1;
+  waveCtx.strokeStyle = waveStroke;
+  waveCtx.beginPath();
+
+  const sliceWidth = (waveCanvas.width / (window.devicePixelRatio || 1)) / bufferLength;
+  let x = 0;
+
+  for (let i = 0; i < bufferLength; i++) {
+    const v = dataArray[i] / 128.0;
+    const y = (v * (waveCanvas.height / (window.devicePixelRatio || 1))) / 2;
+
+    if (i === 0) {
+      waveCtx.moveTo(x, y);
+    } else {
+      waveCtx.lineTo(x, y);
+    }
+
+    x += sliceWidth;
+  }
+
+  waveCtx.lineTo(
+    waveCanvas.width / (window.devicePixelRatio || 1),
+    (waveCanvas.height / (window.devicePixelRatio || 1)) / 2
+  );
+  waveCtx.stroke();
+
+  animationFrame = requestAnimationFrame(drawWaveform);
+}
+
+async function startRecording() {
+  try {
+    try {
+      wakeLock = await navigator.wakeLock.request("screen");
+    } catch (err) {
+      console.log("Error acquiring wake lock.");
+    }
+
+    let stream;
+    try {
+      // Try tab capture first
+      stream = await new Promise((resolve, reject) => {
+        chrome.tabCapture.capture({audio: true}, (s) => {
+          if (s) {
+            resolve(s);
+          } else {
+            reject(new Error('Tab capture failed or not available'));
+          }
+        });
+      });
+      statusText.textContent = "Using tab audio capture.";
+    } catch (tabError) {
+      console.log('Tab capture not available, falling back to microphone', tabError);
+      // Fallback to microphone
+      const audioConstraints = selectedMicrophoneId
+        ? { audio: { deviceId: { exact: selectedMicrophoneId } } }
+        : { audio: true };
+      stream = await navigator.mediaDevices.getUserMedia(audioConstraints);
+      statusText.textContent = "Using microphone audio.";
+    }
+
+    audioContext = new (window.AudioContext || window.webkitAudioContext)();
+    analyser = audioContext.createAnalyser();
+    analyser.fftSize = 256;
+    microphone = audioContext.createMediaStreamSource(stream);
+    microphone.connect(analyser);
+
+    recorder = new MediaRecorder(stream, { mimeType: "audio/webm" });
+    recorder.ondataavailable = (e) => {
+      if (websocket && websocket.readyState === WebSocket.OPEN) {
+        websocket.send(e.data);
+      }
+    };
+    recorder.start(chunkDuration);
+
+    startTime = Date.now();
+    timerInterval = setInterval(updateTimer, 1000);
+    drawWaveform();
+
+    isRecording = true;
+    updateUI();
+  } catch (err) {
+    if (window.location.hostname === "0.0.0.0") {
+      statusText.textContent =
+        "Error accessing audio input. Browsers may block audio access on 0.0.0.0. Try using localhost:8000 instead.";
+    } else {
+      statusText.textContent = "Error accessing audio input. Please check permissions.";
+    }
+    console.error(err);
+  }
+}
+
+async function stopRecording() {
+  if (wakeLock) {
+    try {
+      await wakeLock.release();
+    } catch (e) {
+      // ignore
+    }
+    wakeLock = null;
+  }
+
+  userClosing = true;
+  waitingForStop = true;
+
+  if (websocket && websocket.readyState === WebSocket.OPEN) {
+    const emptyBlob = new Blob([], { type: "audio/webm" });
+    websocket.send(emptyBlob);
+    statusText.textContent = "Recording stopped. Processing final audio...";
+  }
+
+  if (recorder) {
+    recorder.stop();
+    recorder = null;
+  }
+
+  if (microphone) {
+    microphone.disconnect();
+    microphone = null;
+  }
+
+  if (analyser) {
+    analyser = null;
+  }
+
+  if (audioContext && audioContext.state !== "closed") {
+    try {
+      await audioContext.close();
+    } catch (e) {
+      console.warn("Could not close audio context:", e);
+    }
+    audioContext = null;
+  }
+
+  if (animationFrame) {
+    cancelAnimationFrame(animationFrame);
+    animationFrame = null;
+  }
+
+  if (timerInterval) {
+    clearInterval(timerInterval);
+    timerInterval = null;
+  }
+  timerElement.textContent = "00:00";
+  startTime = null;
+
+  isRecording = false;
+  updateUI();
+}
+
+async function toggleRecording() {
+  if (!isRecording) {
+    if (waitingForStop) {
+      console.log("Waiting for stop, early return");
+      return;
+    }
+    console.log("Connecting to WebSocket");
+    try {
+      if (websocket && websocket.readyState === WebSocket.OPEN) {
+        await startRecording();
+      } else {
+        await setupWebSocket();
+        await startRecording();
+      }
+    } catch (err) {
+      statusText.textContent = "Could not connect to WebSocket or access mic. Aborted.";
+      console.error(err);
+    }
+  } else {
+    console.log("Stopping recording");
+    stopRecording();
+  }
+}
+
+function updateUI() {
+  recordButton.classList.toggle("recording", isRecording);
+  recordButton.disabled = waitingForStop;
+
+  if (waitingForStop) {
+    if (statusText.textContent !== "Recording stopped. Processing final audio...") {
+      statusText.textContent = "Please wait for processing to complete...";
+    }
+  } else if (isRecording) {
+    statusText.textContent = "Recording...";
+  } else {
+    if (
+      statusText.textContent !== "Finished processing audio! Ready to record again." &&
+      statusText.textContent !== "Processing finalized or connection closed."
+    ) {
+      statusText.textContent = "Click to start transcription";
+    }
+  }
+  if (!waitingForStop) {
+    recordButton.disabled = false;
+  }
+}
+
+recordButton.addEventListener("click", toggleRecording);
+
+if (microphoneSelect) {
+  microphoneSelect.addEventListener("change", handleMicrophoneChange);
+}
+
+// Settings toggle functionality
+settingsToggle.addEventListener("click", () => {
+  settingsDiv.classList.toggle("visible");
+  settingsToggle.classList.toggle("active");
+});
+
+document.addEventListener('DOMContentLoaded', async () => {
+  try {
+    await enumerateMicrophones();
+  } catch (error) {
+    console.log("Could not enumerate microphones on load:", error);
+  }
+});
+navigator.mediaDevices.addEventListener('devicechange', async () => {
+  console.log('Device change detected, re-enumerating microphones');
+  try {
+    await enumerateMicrophones();
+  } catch (error) {
+    console.log("Error re-enumerating microphones:", error);
+  }
+});
+
+
+async function run() {
+  const micPermission = await navigator.permissions.query({
+    name: "microphone",
+  });
+
+  document.getElementById(
+    "audioPermission"
+  ).innerText = `MICROPHONE: ${micPermission.state}`;
+
+  if (micPermission.state !== "granted") {
+    chrome.tabs.create({ url: "welcome.html" });
+  }
+
+  const intervalId = setInterval(async () => {
+    const micPermission = await navigator.permissions.query({
+      name: "microphone",
+    });
+    if (micPermission.state === "granted") {
+      document.getElementById(
+        "audioPermission"
+      ).innerText = `MICROPHONE: ${micPermission.state}`;
+      clearInterval(intervalId);
+    }
+  }, 100);
+}
+
+void run();

chrome-extension/manifest.json

@@ -0,0 +1,37 @@
+{
+    "manifest_version": 3,
+    "name": "WhisperLiveKit Tab Capture",
+    "version": "1.0",
+    "description": "Capture and transcribe audio from browser tabs using WhisperLiveKit.",
+    "background": {
+        "service_worker": "background.js"
+    },
+    "icons": {
+        "16": "icons/icon16.png",
+        "32": "icons/icon32.png",
+        "48": "icons/icon48.png",
+        "128": "icons/icon128.png"
+    },
+    "action": {
+        "default_title": "WhisperLiveKit Tab Capture",
+        "default_popup": "popup.html"
+    },
+    "permissions": [
+        "scripting",
+        "tabCapture",
+        "offscreen",
+        "activeTab",
+        "storage"
+    ],
+    "web_accessible_resources": [
+        {
+            "resources": [
+                "requestPermissions.html",
+                "requestPermissions.js"
+            ],
+            "matches": [
+                "<all_urls>"
+            ]
+        }
+    ]
+}

chrome-extension/popup.html

@@ -0,0 +1,78 @@
+<!DOCTYPE html>
+<html lang="en">
+
+<head>
+    <meta charset="UTF-8" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <title>WhisperLiveKit</title>
+    <link rel="stylesheet" href="/web/live_transcription.css" />
+</head>
+
+<body>
+    <div class="settings-container">
+        <button id="recordButton">
+            <div class="shape-container">
+                <div class="shape"></div>
+            </div>
+            <div class="recording-info">
+                <div class="wave-container">
+                    <canvas id="waveCanvas"></canvas>
+                </div>
+                <div class="timer">00:00</div>
+            </div>
+        </button>
+
+        <button id="settingsToggle" class="settings-toggle" title="Show/hide settings">
+            <img src="/web/src/settings.svg" alt="Settings" />
+        </button>
+
+        <div class="settings">
+            <div class="field">
+                <label for="websocketInput">Websocket URL</label>
+                <input id="websocketInput" type="text" placeholder="ws://host:port/asr" />
+            </div>
+
+            <div class="field">
+                <label id="microphoneSelectLabel" for="microphoneSelect">Select Microphone</label>
+                <select id="microphoneSelect">
+                    <option value="">Default Microphone</option>
+                </select>
+                        <div id="audioPermission"></div>
+
+            </div>
+
+            <div class="theme-selector-container">
+                <div class="segmented" role="radiogroup" aria-label="Theme selector">
+                    <input type="radio" id="theme-system" name="theme" value="system" />
+                    <label for="theme-system" title="System">
+                        <img src="/web/src/system_mode.svg" alt="" />
+                        <!-- <span>System</span> -->
+                    </label>
+
+                    <input type="radio" id="theme-light" name="theme" value="light" />
+                    <label for="theme-light" title="Light">
+                        <img src="/web/src/light_mode.svg" alt="" />
+                        <!-- <span>Light</span> -->
+                    </label>
+
+                    <input type="radio" id="theme-dark" name="theme" value="dark" />
+                    <label for="theme-dark" title="Dark">
+                        <img src="/web/src/dark_mode.svg" alt="" />
+                        <!-- <span>Dark</span> -->
+                    </label>
+                </div>
+            </div>
+
+        </div>
+    </div>
+
+
+
+    <p id="status"></p>
+
+    <div id="linesTranscript"></div>
+
+    <script src="live_transcription.js"></script>
+</body>
+
+</html>

chrome-extension/requestPermissions.html

@@ -0,0 +1,12 @@
+<!DOCTYPE html>
+<html>
+  <head>
+    <title>Request Permissions</title>
+    <script src="requestPermissions.js"></script>
+  </head>
+  <body>
+    This page exists to workaround an issue with Chrome that blocks permission
+    requests from chrome extensions
+    <button id="requestMicrophone">Request Microphone</button>
+  </body>
+</html>

chrome-extension/requestPermissions.js

@@ -0,0 +1,17 @@
+/**
+ * Requests user permission for microphone access.
+ * @returns {Promise<void>} A Promise that resolves when permission is granted or rejects with an error.
+ */
+async function getUserPermission() {
+  console.log("Getting user permission for microphone access...");
+  await navigator.mediaDevices.getUserMedia({ audio: true });
+  const micPermission = await navigator.permissions.query({
+    name: "microphone",
+  });
+  if (micPermission.state == "granted") {
+    window.close();
+  }
+}
+
+// Call the function to request microphone permission
+getUserPermission();

chrome-extension/sidepanel.js

@@ -0,0 +1,29 @@
+console.log("sidepanel.js");
+
+async function run() {
+  const micPermission = await navigator.permissions.query({
+    name: "microphone",
+  });
+
+  document.getElementById(
+    "audioPermission"
+  ).innerText = `MICROPHONE: ${micPermission.state}`;
+
+  if (micPermission.state !== "granted") {
+    chrome.tabs.create({ url: "requestPermissions.html" });
+  }
+
+  const intervalId = setInterval(async () => {
+    const micPermission = await navigator.permissions.query({
+      name: "microphone",
+    });
+    if (micPermission.state === "granted") {
+      document.getElementById(
+        "audioPermission"
+      ).innerText = `MICROPHONE: ${micPermission.state}`;
+      clearInterval(intervalId);
+    }
+  }, 100);
+}
+
+void run();

chrome-extension/web/live_transcription.css

@@ -0,0 +1,539 @@
+:root {
+  --bg: #ffffff;
+  --text: #111111;
+  --muted: #666666;
+  --border: #e5e5e5;
+  --chip-bg: rgba(0, 0, 0, 0.04);
+  --chip-text: #000000;
+  --spinner-border: #8d8d8d5c;
+  --spinner-top: #b0b0b0;
+  --silence-bg: #f3f3f3;
+  --loading-bg: rgba(255, 77, 77, 0.06);
+  --button-bg: #ffffff;
+  --button-border: #e9e9e9;
+  --wave-stroke: #000000;
+  --label-dia-text: #868686;
+  --label-trans-text: #111111;
+}
+
+@media (prefers-color-scheme: dark) {
+  :root:not([data-theme="light"]) {
+    --bg: #0b0b0b;
+    --text: #e6e6e6;
+    --muted: #9aa0a6;
+    --border: #333333;
+    --chip-bg: rgba(255, 255, 255, 0.08);
+    --chip-text: #e6e6e6;
+    --spinner-border: #555555;
+    --spinner-top: #dddddd;
+    --silence-bg: #1a1a1a;
+    --loading-bg: rgba(255, 77, 77, 0.12);
+    --button-bg: #111111;
+    --button-border: #333333;
+    --wave-stroke: #e6e6e6;
+    --label-dia-text: #b3b3b3;
+    --label-trans-text: #ffffff;
+  }
+}
+
+:root[data-theme="dark"] {
+  --bg: #0b0b0b;
+  --text: #e6e6e6;
+  --muted: #9aa0a6;
+  --border: #333333;
+  --chip-bg: rgba(255, 255, 255, 0.08);
+  --chip-text: #e6e6e6;
+  --spinner-border: #555555;
+  --spinner-top: #dddddd;
+  --silence-bg: #1a1a1a;
+  --loading-bg: rgba(255, 77, 77, 0.12);
+  --button-bg: #111111;
+  --button-border: #333333;
+  --wave-stroke: #e6e6e6;
+  --label-dia-text: #b3b3b3;
+  --label-trans-text: #ffffff;
+}
+
+:root[data-theme="light"] {
+  --bg: #ffffff;
+  --text: #111111;
+  --muted: #666666;
+  --border: #e5e5e5;
+  --chip-bg: rgba(0, 0, 0, 0.04);
+  --chip-text: #000000;
+  --spinner-border: #8d8d8d5c;
+  --spinner-top: #b0b0b0;
+  --silence-bg: #f3f3f3;
+  --loading-bg: rgba(255, 77, 77, 0.06);
+  --button-bg: #ffffff;
+  --button-border: #e9e9e9;
+  --wave-stroke: #000000;
+  --label-dia-text: #868686;
+  --label-trans-text: #111111;
+}
+
+body {
+  font-family: ui-sans-serif, system-ui, sans-serif, 'Apple Color Emoji', 'Segoe UI Emoji', 'Segoe UI Symbol', 'Noto Color Emoji';
+  margin: 20px;
+  text-align: center;
+  background-color: var(--bg);
+  color: var(--text);
+}
+
+.settings-toggle {
+  margin-top: 4px;
+  width: 40px;
+  height: 40px;
+  border: none;
+  border-radius: 50%;
+  background-color: var(--button-bg);
+  cursor: pointer;
+  transition: all 0.3s ease;
+  /* border: 1px solid var(--button-border); */
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  position: relative;
+}
+
+.settings-toggle:hover {
+  background-color: var(--chip-bg);
+}
+
+.settings-toggle img {
+  width: 24px;
+  height: 24px;
+  opacity: 0.7;
+  transition: opacity 0.2s ease, transform 0.3s ease;
+}
+
+.settings-toggle:hover img {
+  opacity: 1;
+}
+
+.settings-toggle.active img {
+  transform: rotate(80deg);
+}
+
+/* Record button */
+#recordButton {
+  width: 50px;
+  height: 50px;
+  border: none;
+  border-radius: 50%;
+  background-color: var(--button-bg);
+  cursor: pointer;
+  transition: all 0.3s ease;
+  border: 1px solid var(--button-border);
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  position: relative;
+}
+
+#recordButton.recording {
+  width: 180px;
+  border-radius: 40px;
+  justify-content: flex-start;
+  padding-left: 20px;
+}
+
+#recordButton:active {
+  transform: scale(0.95);
+}
+
+.shape-container {
+  width: 25px;
+  height: 25px;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+  flex-shrink: 0;
+}
+
+.shape {
+  width: 25px;
+  height: 25px;
+  background-color: rgb(209, 61, 53);
+  border-radius: 50%;
+  transition: all 0.3s ease;
+}
+
+#recordButton:disabled .shape {
+  background-color: #6e6d6d;
+}
+
+#recordButton.recording .shape {
+  border-radius: 5px;
+  width: 25px;
+  height: 25px;
+}
+
+/* Recording elements */
+.recording-info {
+  display: none;
+  align-items: center;
+  margin-left: 15px;
+  flex-grow: 1;
+}
+
+#recordButton.recording .recording-info {
+  display: flex;
+}
+
+.wave-container {
+  width: 60px;
+  height: 30px;
+  position: relative;
+  display: flex;
+  align-items: center;
+  justify-content: center;
+}
+
+#waveCanvas {
+  width: 100%;
+  height: 100%;
+}
+
+.timer {
+  font-size: 14px;
+  font-weight: 500;
+  color: var(--text);
+  margin-left: 10px;
+}
+
+#status {
+  margin-top: 20px;
+  font-size: 16px;
+  color: var(--text);
+}
+
+/* Settings */
+.settings-container {
+  display: flex;
+  justify-content: center;
+  align-items: flex-start;
+  gap: 15px;
+  margin-top: 20px;
+  flex-wrap: wrap;
+}
+
+.settings {
+  display: none;
+  flex-wrap: wrap;
+  align-items: flex-start;
+  gap: 12px;
+  transition: opacity 0.3s ease;
+}
+
+.settings.visible {
+  display: flex;
+}
+
+.field {
+  display: flex;
+  flex-direction: column;
+  align-items: flex-start;
+  gap: 3px;
+}
+
+#chunkSelector,
+#websocketInput,
+#themeSelector,
+#microphoneSelect {
+  font-size: 16px;
+  padding: 5px 8px;
+  border-radius: 8px;
+  border: 1px solid var(--border);
+  background-color: var(--button-bg);
+  color: var(--text);
+  max-height: 30px;
+}
+
+#microphoneSelect {
+  width: 100%;
+  max-width: 190px;
+  min-width: 120px;
+}
+
+#chunkSelector:focus,
+#websocketInput:focus,
+#themeSelector:focus,
+#microphoneSelect:focus {
+  outline: none;
+  border-color: #007bff;
+  box-shadow: 0 0 0 3px rgba(0, 123, 255, 0.15);
+}
+
+label {
+  font-size: 13px;
+  color: var(--muted);
+}
+
+.ws-default {
+  font-size: 12px;
+  color: var(--muted);
+}
+
+/* Segmented pill control for Theme */
+.segmented {
+  display: inline-flex;
+  align-items: stretch;
+  border: 1px solid var(--button-border);
+  background-color: var(--button-bg);
+  border-radius: 999px;
+  overflow: hidden;
+}
+
+.segmented input[type="radio"] {
+  position: absolute;
+  opacity: 0;
+  pointer-events: none;
+}
+
+.theme-selector-container {
+  display: flex;
+  align-items: center;
+  margin-top: 17px;
+}
+
+.segmented label {
+  display: inline-flex;
+  align-items: center;
+  gap: 6px;
+  padding: 6px 12px;
+  font-size: 14px;
+  color: var(--muted);
+  cursor: pointer;
+  user-select: none;
+  transition: background-color 0.2s ease, color 0.2s ease;
+}
+
+.segmented label span {
+  display: none;
+}
+
+.segmented label:hover span {
+  display: inline;
+}
+
+.segmented label:hover {
+  background-color: var(--chip-bg);
+}
+
+.segmented img {
+  width: 16px;
+  height: 16px;
+}
+
+.segmented input[type="radio"]:checked + label {
+  background-color: var(--chip-bg);
+  color: var(--text);
+}
+
+.segmented input[type="radio"]:focus-visible + label,
+.segmented input[type="radio"]:focus + label {
+  outline: 2px solid #007bff;
+  outline-offset: 2px;
+  border-radius: 999px;
+}
+
+/* Transcript area */
+#linesTranscript {
+  margin: 20px auto;
+  max-width: 700px;
+  text-align: left;
+  font-size: 16px;
+}
+
+#linesTranscript p {
+  margin: 0px 0;
+}
+
+#linesTranscript strong {
+  color: var(--text);
+}
+
+#speaker {
+  border: 1px solid var(--border);
+  border-radius: 100px;
+  padding: 2px 10px;
+  font-size: 14px;
+  margin-bottom: 0px;
+}
+
+.label_diarization {
+  background-color: var(--chip-bg);
+  border-radius: 8px 8px 8px 8px;
+  padding: 2px 10px;
+  margin-left: 10px;
+  display: inline-block;
+  white-space: nowrap;
+  font-size: 14px;
+  margin-bottom: 0px;
+  color: var(--label-dia-text);
+}
+
+.label_transcription {
+  background-color: var(--chip-bg);
+  border-radius: 8px 8px 8px 8px;
+  padding: 2px 10px;
+  display: inline-block;
+  white-space: nowrap;
+  margin-left: 10px;
+  font-size: 14px;
+  margin-bottom: 0px;
+  color: var(--label-trans-text);
+}
+
+#timeInfo {
+  color: var(--muted);
+  margin-left: 10px;
+}
+
+.textcontent {
+  font-size: 16px;
+  padding-left: 10px;
+  margin-bottom: 10px;
+  margin-top: 1px;
+  padding-top: 5px;
+  border-radius: 0px 0px 0px 10px;
+}
+
+.buffer_diarization {
+  color: var(--label-dia-text);
+  margin-left: 4px;
+}
+
+.buffer_transcription {
+  color: #7474748c;
+  margin-left: 4px;
+}
+
+.spinner {
+  display: inline-block;
+  width: 8px;
+  height: 8px;
+  border: 2px solid var(--spinner-border);
+  border-top: 2px solid var(--spinner-top);
+  border-radius: 50%;
+  animation: spin 0.7s linear infinite;
+  vertical-align: middle;
+  margin-bottom: 2px;
+  margin-right: 5px;
+}
+
+@keyframes spin {
+  to {
+    transform: rotate(360deg);
+  }
+}
+
+.silence {
+  color: var(--muted);
+  background-color: var(--silence-bg);
+  font-size: 13px;
+  border-radius: 30px;
+  padding: 2px 10px;
+}
+
+.loading {
+  color: var(--muted);
+  background-color: var(--loading-bg);
+  border-radius: 8px 8px 8px 0px;
+  padding: 2px 10px;
+  font-size: 14px;
+  margin-bottom: 0px;
+}
+
+/* for smaller screens */
+/* @media (max-width: 450px) {
+  .settings-container {
+    flex-direction: column;
+    gap: 10px;
+    align-items: center;
+  }
+
+  .settings {
+    justify-content: center;
+    gap: 8px;
+    width: 100%;
+  }
+
+  .field {
+    align-items: center;
+    width: 100%;
+  }
+
+  #websocketInput,
+  #microphoneSelect {
+    min-width: 200px;
+    max-width: 100%;
+  }
+
+  .theme-selector-container {
+    margin-top: 10px;
+  }
+} */
+
+/* @media (max-width: 768px) and (min-width: 451px) {
+  .settings-container {
+    gap: 10px;
+  }
+
+  .settings {
+    gap: 8px;
+  }
+
+  #websocketInput,
+  #microphoneSelect {
+    min-width: 150px;
+    max-width: 300px;
+  }
+} */
+
+/* @media (max-width: 480px) {
+  body {
+    margin: 10px;
+  }
+
+  .settings-toggle {
+    width: 35px;
+    height: 35px;
+  }
+
+  .settings-toggle img {
+    width: 20px;
+    height: 20px;
+  }
+
+  .settings {
+    flex-direction: column;
+    align-items: center;
+    gap: 6px;
+  }
+
+  #websocketInput,
+  #microphoneSelect {
+    max-width: 400px;
+  }
+
+  .segmented label {
+    padding: 4px 8px;
+    font-size: 12px;
+  }
+
+  .segmented img {
+    width: 14px;
+    height: 14px;
+  }
+} */
+
+
+html
+{
+    width: 400px;  /* max: 800px */
+    height: 600px; /* max: 600px */
+    border-radius: 10px;
+
+}

chrome-extension/web/src/dark_mode.svg

@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" height="24px" viewBox="0 -960 960 960" width="24px" fill="#5f6368"><path d="M480-120q-151 0-255.5-104.5T120-480q0-138 90-239.5T440-838q13-2 23 3.5t16 14.5q6 9 6.5 21t-7.5 23q-17 26-25.5 55t-8.5 61q0 90 63 153t153 63q31 0 61.5-9t54.5-25q11-7 22.5-6.5T819-479q10 5 15.5 15t3.5 24q-14 138-117.5 229T480-120Zm0-80q88 0 158-48.5T740-375q-20 5-40 8t-40 3q-123 0-209.5-86.5T364-660q0-20 3-40t8-40q-78 32-126.5 102T200-480q0 116 82 198t198 82Zm-10-270Z"/></svg>

chrome-extension/web/src/light_mode.svg

@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" height="24px" viewBox="0 -960 960 960" width="24px" fill="#5f6368"><path d="M480-360q50 0 85-35t35-85q0-50-35-85t-85-35q-50 0-85 35t-35 85q0 50 35 85t85 35Zm0 80q-83 0-141.5-58.5T280-480q0-83 58.5-141.5T480-680q83 0 141.5 58.5T680-480q0 83-58.5 141.5T480-280ZM80-440q-17 0-28.5-11.5T40-480q0-17 11.5-28.5T80-520h80q17 0 28.5 11.5T200-480q0 17-11.5 28.5T160-440H80Zm720 0q-17 0-28.5-11.5T760-480q0-17 11.5-28.5T800-520h80q17 0 28.5 11.5T920-480q0 17-11.5 28.5T880-440h-80ZM480-760q-17 0-28.5-11.5T440-800v-80q0-17 11.5-28.5T480-920q17 0 28.5 11.5T520-880v80q0 17-11.5 28.5T480-760Zm0 720q-17 0-28.5-11.5T440-80v-80q0-17 11.5-28.5T480-200q17 0 28.5 11.5T520-160v80q0 17-11.5 28.5T480-40ZM226-678l-43-42q-12-11-11.5-28t11.5-29q12-12 29-12t28 12l42 43q11 12 11 28t-11 28q-11 12-27.5 11.5T226-678Zm494 495-42-43q-11-12-11-28.5t11-27.5q11-12 27.5-11.5T734-282l43 42q12 11 11.5 28T777-183q-12 12-29 12t-28-12Zm-42-495q-12-11-11.5-27.5T678-734l42-43q11-12 28-11.5t29 11.5q12 12 12 29t-12 28l-43 42q-12 11-28 11t-28-11ZM183-183q-12-12-12-29t12-28l43-42q12-11 28.5-11t27.5 11q12 11 11.5 27.5T282-226l-42 43q-11 12-28 11.5T183-183Zm297-297Z"/></svg>

chrome-extension/web/src/settings.svg

@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" height="24px" viewBox="0 -960 960 960" width="24px" fill="#5f6368"><path d="M433-80q-27 0-46.5-18T363-142l-9-66q-13-5-24.5-12T307-235l-62 26q-25 11-50 2t-39-32l-47-82q-14-23-8-49t27-43l53-40q-1-7-1-13.5v-27q0-6.5 1-13.5l-53-40q-21-17-27-43t8-49l47-82q14-23 39-32t50 2l62 26q11-8 23-15t24-12l9-66q4-26 23.5-44t46.5-18h94q27 0 46.5 18t23.5 44l9 66q13 5 24.5 12t22.5 15l62-26q25-11 50-2t39 32l47 82q14 23 8 49t-27 43l-53 40q1 7 1 13.5v27q0 6.5-2 13.5l53 40q21 17 27 43t-8 49l-48 82q-14 23-39 32t-50-2l-60-26q-11 8-23 15t-24 12l-9 66q-4 26-23.5 44T527-80h-94Zm7-80h79l14-106q31-8 57.5-23.5T639-327l99 41 39-68-86-65q5-14 7-29.5t2-31.5q0-16-2-31.5t-7-29.5l86-65-39-68-99 42q-22-23-48.5-38.5T533-694l-13-106h-79l-14 106q-31 8-57.5 23.5T321-633l-99-41-39 68 86 64q-5 15-7 30t-2 32q0 16 2 31t7 30l-86 65 39 68 99-42q22 23 48.5 38.5T427-266l13 106Zm42-180q58 0 99-41t41-99q0-58-41-99t-99-41q-59 0-99.5 41T342-480q0 58 40.5 99t99.5 41Zm-2-140Z"/></svg>

chrome-extension/web/src/system_mode.svg

@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" height="24px" viewBox="0 -960 960 960" width="24px" fill="#5f6368"><path d="M396-396q-32-32-58.5-67T289-537q-5 14-6.5 28.5T281-480q0 83 58 141t141 58q14 0 28.5-2t28.5-6q-39-22-74-48.5T396-396Zm85 196q-56 0-107-21t-91-61q-40-40-61-91t-21-107q0-51 17-97.5t50-84.5q13-14 32-9.5t27 24.5q21 55 52.5 104t73.5 91q42 42 91 73.5T648-326q20 8 24.5 27t-9.5 32q-38 33-84.5 50T481-200Zm223-192q-16-5-23-20.5t-4-32.5q9-48-6-94.5T621-621q-35-35-80.5-49.5T448-677q-17 3-32-4t-21-23q-6-16 1.5-31t23.5-19q69-15 138 4.5T679-678q51 51 71 120t5 138q-4 17-19 25t-32 3ZM480-840q-17 0-28.5-11.5T440-880v-40q0-17 11.5-28.5T480-960q17 0 28.5 11.5T520-920v40q0 17-11.5 28.5T480-840Zm0 840q-17 0-28.5-11.5T440-40v-40q0-17 11.5-28.5T480-120q17 0 28.5 11.5T520-80v40q0 17-11.5 28.5T480 0Zm255-734q-12-12-12-28.5t12-28.5l28-28q11-11 27.5-11t28.5 11q12 12 12 28.5T819-762l-28 28q-12 12-28 12t-28-12ZM141-141q-12-12-12-28.5t12-28.5l28-28q12-12 28-12t28 12q12 12 12 28.5T225-169l-28 28q-11 11-27.5 11T141-141Zm739-299q-17 0-28.5-11.5T840-480q0-17 11.5-28.5T880-520h40q17 0 28.5 11.5T960-480q0 17-11.5 28.5T920-440h-40Zm-840 0q-17 0-28.5-11.5T0-480q0-17 11.5-28.5T40-520h40q17 0 28.5 11.5T120-480q0 17-11.5 28.5T80-440H40Zm779 299q-12 12-28.5 12T762-141l-28-28q-12-12-12-28t12-28q12-12 28.5-12t28.5 12l28 28q11 11 11 27.5T819-141ZM226-735q-12 12-28.5 12T169-735l-28-28q-11-11-11-27.5t11-28.5q12-12 28.5-12t28.5 12l28 28q12 12 12 28t-12 28Zm170 339Z"/></svg>

chrome-extension/welcome.html

@@ -0,0 +1,12 @@
+<!DOCTYPE html>
+<html>
+  <head>
+    <title>Welcome</title>
+    <script src="welcome.js"></script>
+  </head>
+  <body>
+    This page exists to workaround an issue with Chrome that blocks permission
+    requests from chrome extensions
+    <!-- <button id="requestMicrophone">Request Microphone</button> -->
+  </body>
+</html>

pyproject.toml

@@ -4,8 +4,8 @@ build-backend = "setuptools.build_meta"
 
 [project]
 name = "whisperlivekit"
-version = "0.2.6"
+version = "0.2.9"
-description = "Real-time, Fully Local Whisper's Speech-to-Text and Speaker Diarization"
+description = "Real-time speech-to-text with speaker diarization using Whisper"
 readme = "README.md"
 authors = [
     { name = "Quentin Fuxa" }
@@ -18,6 +18,11 @@ classifiers = [
     "License :: OSI Approved :: MIT License",
     "Programming Language :: Python :: 3.9",
     "Programming Language :: Python :: 3.10",
+    "Programming Language :: Python :: 3.11",
+    "Programming Language :: Python :: 3.12",
+    "Programming Language :: Python :: 3.13",
+    "Programming Language :: Python :: 3.14",
+    "Programming Language :: Python :: 3.15",
     "Topic :: Scientific/Engineering :: Artificial Intelligence",
     "Topic :: Multimedia :: Sound/Audio :: Speech"
 ]
@@ -28,19 +33,15 @@ dependencies = [
     "faster-whisper",
     "uvicorn",
     "websockets",
-    "torch",
+    "torchaudio>=2.0.0",
+    "torch>=2.0.0",
     "tqdm",
     "tiktoken",
-    'triton>=2.0.0,<3; platform_machine == "x86_64" and (sys_platform == "linux" or sys_platform == "linux2")'
+    'triton>=2.0.0; platform_machine == "x86_64" and (sys_platform == "linux" or sys_platform == "linux2")'
 ]
 
 [project.optional-dependencies]
-diarization = ["diart"]
 sentence = ["mosestokenizer", "wtpsplit"]
-whisper = ["whisper"]
-whisper-timestamped = ["whisper-timestamped"]
-mlx-whisper = ["mlx-whisper"]
-openai = ["openai"]
 
 [project.urls]
 Homepage = "https://github.com/QuentinFuxa/WhisperLiveKit"

whisperlivekit/__init__.py

@@ -1,12 +1,13 @@
 from .audio_processor import AudioProcessor
 from .core import TranscriptionEngine
 from .parse_args import parse_args
-from .web.web_interface import get_web_interface_html
+from .web.web_interface import get_web_interface_html, get_inline_ui_html
 
 __all__ = [
     "TranscriptionEngine",
     "AudioProcessor",
     "parse_args",
     "get_web_interface_html",
+    "get_inline_ui_html",
     "download_simulstreaming_backend",
 ]

whisperlivekit/audio_processor.py

@@ -4,13 +4,11 @@ from time import time, sleep
 import math
 import logging
 import traceback
-from datetime import timedelta
+from whisperlivekit.timed_objects import ASRToken, Silence, Line
-from whisperlivekit.timed_objects import ASRToken, Silence
+from whisperlivekit.core import TranscriptionEngine, online_factory, online_diarization_factory, online_translation_factory
-from whisperlivekit.core import TranscriptionEngine, online_factory
 from whisperlivekit.ffmpeg_manager import FFmpegManager, FFmpegState
-from whisperlivekit.remove_silences import handle_silences
-from whisperlivekit.trail_repetition import trim_tail_repetition
 from whisperlivekit.silero_vad_iterator import FixedVADIterator
+from whisperlivekit.results_formater import format_output
 # Set up logging once
 logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s")
 logger = logging.getLogger(__name__)
@@ -18,9 +16,16 @@ logger.setLevel(logging.DEBUG)
 
 SENTINEL = object() # unique sentinel object for end of stream marker
 
-def format_time(seconds: float) -> str:
+
-    """Format seconds as HH:MM:SS."""
+async def get_all_from_queue(queue):
-    return str(timedelta(seconds=int(seconds)))
+    items = []
+    try:
+        while True:
+            item = queue.get_nowait()
+            items.append(item)
+    except asyncio.QueueEmpty:
+        pass
+    return items
 
 class AudioProcessor:
     """
@@ -47,6 +52,7 @@ class AudioProcessor:
         self.last_ffmpeg_activity = time()
         self.ffmpeg_health_check_interval = 5
         self.ffmpeg_max_idle_time = 10
+        self.is_pcm_input = self.args.pcm_input
         self.debug = False
 
         # State management
@@ -54,6 +60,7 @@ class AudioProcessor:
         self.silence = False
         self.silence_duration = 0.0
         self.tokens = []
+        self.translated_segments = []
         self.buffer_transcription = ""
         self.buffer_diarization = ""
         self.end_buffer = 0
@@ -66,7 +73,6 @@ class AudioProcessor:
         # Models and processing
         self.asr = models.asr
         self.tokenizer = models.tokenizer
-        self.diarization = models.diarization
         self.vac_model = models.vac_model
         if self.args.vac:
             self.vac = FixedVADIterator(models.vac_model)
@@ -87,18 +93,21 @@ class AudioProcessor:
         
         self.transcription_queue = asyncio.Queue() if self.args.transcription else None
         self.diarization_queue = asyncio.Queue() if self.args.diarization else None
+        self.translation_queue = asyncio.Queue() if self.args.target_language else None
         self.pcm_buffer = bytearray()
 
-        # Task references
         self.transcription_task = None
         self.diarization_task = None
         self.ffmpeg_reader_task = None
         self.watchdog_task = None
         self.all_tasks_for_cleanup = []
         
-        # Initialize transcription engine if enabled
         if self.args.transcription:
-            self.online = online_factory(self.args, models.asr, models.tokenizer)
+            self.online = online_factory(self.args, models.asr, models.tokenizer)            
+        if self.args.diarization:
+            self.diarization = online_diarization_factory(self.args, models.diarization_model)
+        if self.args.target_language:
+            self.online_translation = online_translation_factory(self.args, models.translation_model)
 
     def convert_pcm_to_float(self, pcm_buffer):
         """Convert PCM buffer in s16le format to normalized NumPy array."""
@@ -108,17 +117,6 @@ class AudioProcessor:
         """Thread-safe update of transcription with new data."""
         async with self.lock:
             self.tokens.extend(new_tokens)
-            
-            # self.tokens, has_been_trimmed = trim_tail_repetition(
-            #     self.tokens,
-            #     key=lambda t: t.text.strip().lower(),
-            #     min_block=2,        # avoid trimming single '.' loops; set to 1 if you want to remove those too
-            #     max_tail=200,
-            #     prefer="longest",   # prefer removing the longest repeated phrase
-            #     keep=1
-            # )
-            # if has_been_trimmed:
-            #     print('HAS BEEN TRIMMED !')
             self.buffer_transcription = buffer
             self.end_buffer = end_buffer
             self.sep = sep
@@ -133,7 +131,7 @@ class AudioProcessor:
     async def add_dummy_token(self):
         """Placeholder token when no transcription is available."""
         async with self.lock:
-            current_time = time() - self.beg_loop
+            current_time = time() - self.beg_loop if self.beg_loop else 0
             self.tokens.append(ASRToken(
                 start=current_time, end=current_time + 1,
                 text=".", speaker=-1, is_dummy=True
@@ -156,6 +154,7 @@ class AudioProcessor:
                 
             return {
                 "tokens": self.tokens.copy(),
+                "translated_segments": self.translated_segments.copy(),
                 "buffer_transcription": self.buffer_transcription,
                 "buffer_diarization": self.buffer_diarization,
                 "end_buffer": self.end_buffer,
@@ -169,6 +168,7 @@ class AudioProcessor:
         """Reset all state variables to initial values."""
         async with self.lock:
             self.tokens = []
+            self.translated_segments = []
             self.buffer_transcription = self.buffer_diarization = ""
             self.end_buffer = self.end_attributed_speaker = 0
             self.beg_loop = time()
@@ -209,54 +209,7 @@ class AudioProcessor:
                         continue
                     
                 self.pcm_buffer.extend(chunk)
-
+                await self.handle_pcm_data()
-                # Process when enough data
-                if len(self.pcm_buffer) >= self.bytes_per_sec:
-                    if len(self.pcm_buffer) > self.max_bytes_per_sec:
-                        logger.warning(
-                            f"Audio buffer too large: {len(self.pcm_buffer) / self.bytes_per_sec:.2f}s. "
-                            f"Consider using a smaller model."
-                        )
-
-                    # Process audio chunk
-                    pcm_array = self.convert_pcm_to_float(self.pcm_buffer[:self.max_bytes_per_sec])
-                    self.pcm_buffer = self.pcm_buffer[self.max_bytes_per_sec:]
-                    
-                    res = None
-                    end_of_audio = False
-                    silence_buffer = None
-                    
-                    if self.args.vac:
-                        res = self.vac(pcm_array)
-                    
-                    if res is not None:
-                        if res.get('end', 0) > res.get('start', 0):
-                            end_of_audio = True
-                        elif self.silence: #end of silence
-                            self.silence = False
-                            silence_buffer = Silence(duration=time() - self.start_silence)
-                            
-                    if silence_buffer:
-                        if self.args.transcription and self.transcription_queue:
-                            await self.transcription_queue.put(silence_buffer)
-                        if self.args.diarization and self.diarization_queue:
-                            await self.diarization_queue.put(silence_buffer)
-
-                    if not self.silence:                            
-                        if self.args.transcription and self.transcription_queue:
-                            await self.transcription_queue.put(pcm_array.copy())
-
-                        if self.args.diarization and self.diarization_queue:
-                            await self.diarization_queue.put(pcm_array.copy())
-                        
-                        self.silence_duration = 0.0
-                        if end_of_audio:
-                            self.silence = True
-                            self.start_silence = time()
-
-                    # Sleep if no processing is happening
-                    if not self.args.transcription and not self.args.diarization:
-                        await asyncio.sleep(0.1)
                     
                     
                     
@@ -277,6 +230,8 @@ class AudioProcessor:
         if self.args.diarization and self.diarization_queue:
             await self.diarization_queue.put(SENTINEL)
             logger.debug("Sentinel put into diarization_queue.")
+        if self.args.target_language and self.translation_queue:
+            await self.translation_queue.put(SENTINEL)
 
 
     async def transcription_processor(self):
@@ -303,12 +258,12 @@ class AudioProcessor:
                 if type(item) is Silence:
                     asr_processing_logs += f" + Silence of = {item.duration:.2f}s"
                     if self.tokens:
-                        asr_processing_logs += " | last_end = {self.tokens[-1].end} |"
+                        asr_processing_logs += f" | last_end = {self.tokens[-1].end} |"
                 logger.info(asr_processing_logs)
                 
                 if type(item) is Silence:
                     cumulative_pcm_duration_stream_time += item.duration
-                    self.online.insert_silence(item.duration, self.tokens[-1].end)
+                    self.online.insert_silence(item.duration, self.tokens[-1].end if self.tokens else 0)
                     continue
                 
                 if isinstance(item, np.ndarray):
@@ -320,9 +275,6 @@ class AudioProcessor:
                 cumulative_pcm_duration_stream_time += duration_this_chunk
                 stream_time_end_of_current_pcm = cumulative_pcm_duration_stream_time
 
-                
-                    
-
                 self.online.insert_audio_chunk(pcm_array, stream_time_end_of_current_pcm)
                 new_tokens, current_audio_processed_upto = self.online.process_iter()
                 
@@ -350,6 +302,11 @@ class AudioProcessor:
                 await self.update_transcription(
                     new_tokens, buffer_text, new_end_buffer, self.sep
                 )
+                
+                if new_tokens and self.args.target_language and self.translation_queue:
+                    for token in new_tokens:
+                        await self.translation_queue.put(token)
+                        
                 self.transcription_queue.task_done()
                 
             except Exception as e:
@@ -404,6 +361,49 @@ class AudioProcessor:
                     self.diarization_queue.task_done()
         logger.info("Diarization processor task finished.")
 
+    async def translation_processor(self, online_translation):
+        # the idea is to ignore diarization for the moment. We use only transcription tokens. 
+        # And the speaker is attributed given the segments used for the translation
+        # in the future we want to have different languages for each speaker etc, so it will be more complex.
+        while True:
+            try:
+                token = await self.translation_queue.get() #block until at least 1 token
+                if token is SENTINEL:
+                    logger.debug("Translation processor received sentinel. Finishing.")
+                    self.translation_queue.task_done()
+                    break
+                
+                # get all the available tokens for translation. The more words, the more precise
+                tokens_to_process = [token]
+                additional_tokens = await get_all_from_queue(self.translation_queue)
+                
+                sentinel_found = False
+                for additional_token in additional_tokens:
+                    if additional_token is SENTINEL:
+                        sentinel_found = True
+                        break
+                    tokens_to_process.append(additional_token)                
+                if tokens_to_process:
+                    online_translation.insert_tokens(tokens_to_process)
+                    self.translated_segments = online_translation.process()
+                
+                self.translation_queue.task_done()
+                for _ in additional_tokens:
+                    self.translation_queue.task_done()
+                
+                if sentinel_found:
+                    logger.debug("Translation processor received sentinel in batch. Finishing.")
+                    break
+                
+            except Exception as e:
+                logger.warning(f"Exception in translation_processor: {e}")
+                logger.warning(f"Traceback: {traceback.format_exc()}")
+                if 'token' in locals() and token is not SENTINEL:
+                    self.translation_queue.task_done()
+                if 'additional_tokens' in locals():
+                    for _ in additional_tokens:
+                        self.translation_queue.task_done()
+        logger.info("Translation processor task finished.")
 
     async def results_formatter(self):
         """Format processing results for output."""
@@ -433,7 +433,7 @@ class AudioProcessor:
                 buffer_diarization = state["buffer_diarization"]
                 end_attributed_speaker = state["end_attributed_speaker"]
                 sep = state["sep"]
-                
+                                
                 # Add dummy tokens if needed
                 if (not tokens or tokens[-1].is_dummy) and not self.args.transcription and self.args.diarization:
                     await self.add_dummy_token()
@@ -442,45 +442,13 @@ class AudioProcessor:
                     tokens = state["tokens"]
                 
                 # Format output
-                previous_speaker = -1
+                lines, undiarized_text, buffer_transcription, buffer_diarization = format_output(
-                lines = []
+                    state,
-                last_end_diarized = 0
+                    self.silence,
-                undiarized_text = []
+                    current_time = time() - self.beg_loop if self.beg_loop else None,
-                current_time = time() - self.beg_loop if self.beg_loop else None
+                    args = self.args,
-                tokens, buffer_transcription, buffer_diarization = handle_silences(tokens, buffer_transcription, buffer_diarization, current_time, self.silence)
+                    debug = self.debug
-                for token in tokens:
+                )
-                    speaker = token.speaker
-                    
-                    if speaker == -1: #Speaker -1 means no attributed by diarization. In the frontend, it should appear under 'Speaker 1'
-                        speaker = 1
-                    
-                    # Handle diarization
-                    if self.args.diarization and not tokens[-1].speaker == -2:
-                        if (speaker in [-1, 0]) and token.end >= end_attributed_speaker:
-                            undiarized_text.append(token.text)
-                            continue
-                        elif (speaker in [-1, 0]) and token.end < end_attributed_speaker:
-                            speaker = previous_speaker
-                        if speaker not in [-1, 0]:
-                            last_end_diarized = max(token.end, last_end_diarized)
-
-                    debug_info = ""
-                    if self.debug:
-                        debug_info = f"[{format_time(token.start)} : {format_time(token.end)}]"
-                    if speaker != previous_speaker or not lines:
-                        lines.append({
-                            "speaker": speaker,
-                            "text": token.text + debug_info,
-                            "beg": format_time(token.start),
-                            "end": format_time(token.end),
-                            "diff": round(token.end - last_end_diarized, 2)
-                        })
-                        previous_speaker = speaker
-                    elif token.text:  # Only append if text isn't empty
-                        lines[-1]["text"] += sep + token.text + debug_info
-                        lines[-1]["end"] = format_time(token.end)
-                        lines[-1]["diff"] = round(token.end - last_end_diarized, 2)
-
                 # Handle undiarized text
                 if undiarized_text:
                     combined = sep.join(undiarized_text)
@@ -490,31 +458,27 @@ class AudioProcessor:
                     buffer_diarization = combined
                 
                 response_status = "active_transcription"
-                final_lines_for_response = lines.copy()
-
                 if not tokens and not buffer_transcription and not buffer_diarization:
                     response_status = "no_audio_detected"
-                    final_lines_for_response = []
+                    lines = []
-                elif response_status == "active_transcription" and not final_lines_for_response:
+                elif response_status == "active_transcription" and not lines:
-                    final_lines_for_response = [{
+                    lines = [Line(
-                        "speaker": 1,
+                        speaker=1,
-                        "text": "",
+                        start=state.get("end_buffer", 0),
-                        "beg": format_time(state.get("end_buffer", 0)),
+                        end=state.get("end_buffer", 0)        
-                        "end": format_time(state.get("end_buffer", 0)),
+                    )]
-                        "diff": 0
-                    }]
                 
                 response = {
                     "status": response_status,
-                    "lines": final_lines_for_response,
+                    "lines": [line.to_dict() for line in lines],
                     "buffer_transcription": buffer_transcription,
                     "buffer_diarization": buffer_diarization,
                     "remaining_time_transcription": state["remaining_time_transcription"],
-                    "remaining_time_diarization": state["remaining_time_diarization"]
+                    "remaining_time_diarization": state["remaining_time_diarization"] if self.args.diarization else 0
                 }
                 
                 current_response_signature = f"{response_status} | " + \
-                                           ' '.join([f"{line['speaker']} {line['text']}" for line in final_lines_for_response]) + \
+                                           ' '.join([f"{line.speaker} {line.text}" for line in lines]) + \
                                            f" | {buffer_transcription} | {buffer_diarization}"
                 
                 trans = state["remaining_time_transcription"]
@@ -525,7 +489,7 @@ class AudioProcessor:
                     or round(trans, 1) != round(last_sent_trans, 1)
                     or round(diar, 1) != round(last_sent_diar, 1)
                 )
-                if should_push and (final_lines_for_response or buffer_transcription or buffer_diarization or response_status == "no_audio_detected" or trans > 0 or diar > 0):
+                if should_push and (lines or buffer_transcription or buffer_diarization or response_status == "no_audio_detected" or trans > 0 or diar > 0):
                     yield response
                     self.last_response_content = current_response_signature
                     last_sent_trans = trans
@@ -541,7 +505,6 @@ class AudioProcessor:
                     
                     if all_processors_done:
                         logger.info("Results formatter: All upstream processors are done and in stopping state. Terminating.")
-                        final_state = await self.get_current_state()
                         return
                 
                 await asyncio.sleep(0.1)  # Avoid overwhelming the client
@@ -581,6 +544,11 @@ class AudioProcessor:
             self.all_tasks_for_cleanup.append(self.diarization_task)
             processing_tasks_for_watchdog.append(self.diarization_task)
         
+        if self.args.target_language and self.args.lan != 'auto':
+            self.translation_task = asyncio.create_task(self.translation_processor(self.online_translation))
+            self.all_tasks_for_cleanup.append(self.translation_task)
+            processing_tasks_for_watchdog.append(self.translation_task)
+        
         self.ffmpeg_reader_task = asyncio.create_task(self.ffmpeg_stdout_reader())
         self.all_tasks_for_cleanup.append(self.ffmpeg_reader_task)
         processing_tasks_for_watchdog.append(self.ffmpeg_reader_task)
@@ -623,20 +591,21 @@ class AudioProcessor:
         
     async def cleanup(self):
         """Clean up resources when processing is complete."""
-        logger.info("Starting cleanup of AudioProcessor resources.")        
+        logger.info("Starting cleanup of AudioProcessor resources.")
+        self.is_stopping = True
         for task in self.all_tasks_for_cleanup:
             if task and not task.done():
                 task.cancel()
-        
+            
-        created_tasks = [t for t in self.all_tasks_for_cleanup if t]
+            created_tasks = [t for t in self.all_tasks_for_cleanup if t]
-        if created_tasks:
+            if created_tasks:
-            await asyncio.gather(*created_tasks, return_exceptions=True)
+                await asyncio.gather(*created_tasks, return_exceptions=True)
-        logger.info("All processing tasks cancelled or finished.")
+            logger.info("All processing tasks cancelled or finished.")
-        await self.ffmpeg_manager.stop()
+            await self.ffmpeg_manager.stop()
-        logger.info("FFmpeg manager stopped.")
+            logger.info("FFmpeg manager stopped.")
-        if self.args.diarization and hasattr(self, 'diarization') and hasattr(self.diarization, 'close'):
+            if self.args.diarization and hasattr(self, 'diarization') and hasattr(self.diarization, 'close'):
-            self.diarization.close()
+                self.diarization.close()
-        logger.info("AudioProcessor cleanup complete.")
+            logger.info("AudioProcessor cleanup complete.")
 
 
     async def process_audio(self, message):
@@ -656,10 +625,65 @@ class AudioProcessor:
             logger.warning("AudioProcessor is stopping. Ignoring incoming audio.")
             return
 
-        success = await self.ffmpeg_manager.write_data(message)
+        if self.is_pcm_input:
-        if not success:
+            self.pcm_buffer.extend(message)
-            ffmpeg_state = await self.ffmpeg_manager.get_state()
+            await self.handle_pcm_data()
-            if ffmpeg_state == FFmpegState.FAILED:
+        else:
-                logger.error("FFmpeg is in FAILED state, cannot process audio")
+            success = await self.ffmpeg_manager.write_data(message)
-            else:
+            if not success:
-                logger.warning("Failed to write audio data to FFmpeg")
+                ffmpeg_state = await self.ffmpeg_manager.get_state()
+                if ffmpeg_state == FFmpegState.FAILED:
+                    logger.error("FFmpeg is in FAILED state, cannot process audio")
+                else:
+                    logger.warning("Failed to write audio data to FFmpeg")
+
+    async def handle_pcm_data(self):
+        # Process when enough data
+        if len(self.pcm_buffer) < self.bytes_per_sec:
+            return
+
+        if len(self.pcm_buffer) > self.max_bytes_per_sec:
+            logger.warning(
+                f"Audio buffer too large: {len(self.pcm_buffer) / self.bytes_per_sec:.2f}s. "
+                f"Consider using a smaller model."
+            )
+
+        # Process audio chunk
+        pcm_array = self.convert_pcm_to_float(self.pcm_buffer[:self.max_bytes_per_sec])
+        self.pcm_buffer = self.pcm_buffer[self.max_bytes_per_sec:]
+
+        res = None
+        end_of_audio = False
+        silence_buffer = None
+
+        if self.args.vac:
+            res = self.vac(pcm_array)
+
+        if res is not None:
+            if res.get("end", 0) > res.get("start", 0):
+                end_of_audio = True
+            elif self.silence: #end of silence
+                self.silence = False
+                silence_buffer = Silence(duration=time() - self.start_silence)
+
+        if silence_buffer:
+            if self.args.transcription and self.transcription_queue:
+                await self.transcription_queue.put(silence_buffer)
+            if self.args.diarization and self.diarization_queue:
+                await self.diarization_queue.put(silence_buffer)
+
+        if not self.silence:
+            if self.args.transcription and self.transcription_queue:
+                await self.transcription_queue.put(pcm_array.copy())
+
+            if self.args.diarization and self.diarization_queue:
+                await self.diarization_queue.put(pcm_array.copy())
+
+            self.silence_duration = 0.0
+
+            if end_of_audio:
+                self.silence = True
+                self.start_silence = time()
+
+        if not self.args.transcription and not self.args.diarization:
+            await asyncio.sleep(0.1)

whisperlivekit/basic_server.py

@@ -2,7 +2,7 @@ from contextlib import asynccontextmanager
 from fastapi import FastAPI, WebSocket, WebSocketDisconnect
 from fastapi.responses import HTMLResponse
 from fastapi.middleware.cors import CORSMiddleware
-from whisperlivekit import TranscriptionEngine, AudioProcessor, get_web_interface_html, parse_args
+from whisperlivekit import TranscriptionEngine, AudioProcessor, get_inline_ui_html, parse_args
 import asyncio
 import logging
 from starlette.staticfiles import StaticFiles
@@ -19,6 +19,15 @@ transcription_engine = None
 
 @asynccontextmanager
 async def lifespan(app: FastAPI):
+    
+    #to remove after 0.2.8
+    if args.backend == "simulstreaming" and not args.disable_fast_encoder:
+        logger.warning(f"""
+{'='*50}
+WhisperLiveKit 0.2.8 has introduced a new fast encoder feature using MLX Whisper or Faster Whisper for improved speed. Use --disable-fast-encoder to disable if you encounter issues.
+{'='*50}
+    """)
+    
     global transcription_engine
     transcription_engine = TranscriptionEngine(
         **vars(args),
@@ -38,7 +47,7 @@ app.mount("/web", StaticFiles(directory=str(web_dir)), name="web")
 
 @app.get("/")
 async def get():
-    return HTMLResponse(get_web_interface_html())
+    return HTMLResponse(get_inline_ui_html())
 
 
 async def handle_websocket_results(websocket, results_generator):
@@ -52,7 +61,7 @@ async def handle_websocket_results(websocket, results_generator):
     except WebSocketDisconnect:
         logger.info("WebSocket disconnected while handling results (client likely closed connection).")
     except Exception as e:
-        logger.error(f"Error in WebSocket results handler: {e}")
+        logger.exception(f"Error in WebSocket results handler: {e}")
 
 
 @app.websocket("/asr")

whisperlivekit/core.py

@@ -33,6 +33,7 @@ class TranscriptionEngine:
             "model_dir": None,
             "lan": "auto",
             "task": "transcribe",
+            "target_language": "",
             "backend": "faster-whisper",
             "vac": True,
             "vac_chunk_size": 0.04,
@@ -41,11 +42,13 @@ class TranscriptionEngine:
             "ssl_keyfile": None,
             "transcription": True,
             "vad": True,
+            "pcm_input": False,
             # whisperstreaming params:
             "buffer_trimming": "segment",
             "confidence_validation": False,
             "buffer_trimming_sec": 15,
             # simulstreaming params:
+            "disable_fast_encoder": False,
             "frame_threshold": 25,
             "beams": 1,
             "decoder_type": None,
@@ -57,10 +60,11 @@ class TranscriptionEngine:
             "static_init_prompt": None,
             "max_context_tokens": None,
             "model_path": './base.pt',
-            "diarization_backend": "diart",
+            "diarization_backend": "sortformer",
-            # diart params:
+            # diarization params:
+            "disable_punctuation_split" : False,
             "segmentation_model": "pyannote/segmentation-3.0",
-            "embedding_model": "pyannote/embedding",
+            "embedding_model": "pyannote/embedding",         
         }
 
         config_dict = {**defaults, **kwargs}
@@ -97,7 +101,7 @@ class TranscriptionEngine:
                 simulstreaming_kwargs = {}
                 for attr in ['frame_threshold', 'beams', 'decoder_type', 'audio_max_len', 'audio_min_len', 
                             'cif_ckpt_path', 'never_fire', 'init_prompt', 'static_init_prompt', 
-                            'max_context_tokens', 'model_path', 'warmup_file', 'preload_model_count']:
+                            'max_context_tokens', 'model_path', 'warmup_file', 'preload_model_count', 'disable_fast_encoder']:
                     if hasattr(self.args, attr):
                         simulstreaming_kwargs[attr] = getattr(self.args, attr)
         
@@ -121,15 +125,24 @@ class TranscriptionEngine:
         if self.args.diarization:
             if self.args.diarization_backend == "diart":
                 from whisperlivekit.diarization.diart_backend import DiartDiarization
-                self.diarization = DiartDiarization(
+                self.diarization_model = DiartDiarization(
                     block_duration=self.args.min_chunk_size,
                     segmentation_model_name=self.args.segmentation_model,
                     embedding_model_name=self.args.embedding_model
                 )
             elif self.args.diarization_backend == "sortformer":
-                raise ValueError('Sortformer backend in developement')
+                from whisperlivekit.diarization.sortformer_backend import SortformerDiarization
+                self.diarization_model = SortformerDiarization()
             else:
                 raise ValueError(f"Unknown diarization backend: {self.args.diarization_backend}")
+        
+        self.translation_model = None
+        if self.args.target_language:
+            if self.args.lan == 'auto':
+                raise Exception('Translation cannot be set with language auto')
+            else:
+                from whisperlivekit.translation.translation import load_model
+                self.translation_model = load_model([self.args.lan]) #in the future we want to handle different languages for different speakers
             
         TranscriptionEngine._initialized = True
 
@@ -152,4 +165,22 @@ def online_factory(args, asr, tokenizer, logfile=sys.stderr):
             confidence_validation = args.confidence_validation
         )
     return online
-  
+  
+  
+def online_diarization_factory(args, diarization_backend):
+    if args.diarization_backend == "diart":
+        online = diarization_backend
+        # Not the best here, since several user/instances will share the same backend, but diart is not SOTA anymore and sortformer is recommended
+    
+    if args.diarization_backend == "sortformer":
+        from whisperlivekit.diarization.sortformer_backend import SortformerDiarizationOnline
+        online = SortformerDiarizationOnline(shared_model=diarization_backend)
+    return online
+
+
+def online_translation_factory(args, translation_model):
+    #should be at speaker level in the future:
+    #one shared nllb model for all speaker
+    #one tokenizer per speaker/language
+    from whisperlivekit.translation.translation import OnlineTranslation
+    return OnlineTranslation(translation_model, [args.lan], [args.target_language])

whisperlivekit/diarization/sortformer_backend.py

@@ -1,145 +1,465 @@
 import numpy as np
 import torch
 import logging
+import threading
+import time
+import wave
+from typing import List, Optional
+from queue import SimpleQueue, Empty
+
 from whisperlivekit.timed_objects import SpeakerSegment
 
 logger = logging.getLogger(__name__)
 
 try:
     from nemo.collections.asr.models import SortformerEncLabelModel
+    from nemo.collections.asr.modules import AudioToMelSpectrogramPreprocessor
 except ImportError:
     raise SystemExit("""Please use `pip install "git+https://github.com/NVIDIA/NeMo.git@main#egg=nemo_toolkit[asr]"` to use the Sortformer diarization""")
 
+
+class StreamingSortformerState:
+    """
+    This class creates a class instance that will be used to store the state of the
+    streaming Sortformer model.
+
+    Attributes:
+        spkcache (torch.Tensor): Speaker cache to store embeddings from start
+        spkcache_lengths (torch.Tensor): Lengths of the speaker cache
+        spkcache_preds (torch.Tensor): The speaker predictions for the speaker cache parts
+        fifo (torch.Tensor): FIFO queue to save the embedding from the latest chunks
+        fifo_lengths (torch.Tensor): Lengths of the FIFO queue
+        fifo_preds (torch.Tensor): The speaker predictions for the FIFO queue parts
+        spk_perm (torch.Tensor): Speaker permutation information for the speaker cache
+        mean_sil_emb (torch.Tensor): Mean silence embedding
+        n_sil_frames (torch.Tensor): Number of silence frames
+    """
+
+    def __init__(self):
+        self.spkcache = None  # Speaker cache to store embeddings from start
+        self.spkcache_lengths = None
+        self.spkcache_preds = None  # speaker cache predictions
+        self.fifo = None  # to save the embedding from the latest chunks
+        self.fifo_lengths = None
+        self.fifo_preds = None
+        self.spk_perm = None
+        self.mean_sil_emb = None
+        self.n_sil_frames = None
+
+
 class SortformerDiarization:
-    def __init__(self, model_name="nvidia/diar_streaming_sortformer_4spk-v2"):
+    def __init__(self, model_name: str = "nvidia/diar_streaming_sortformer_4spk-v2"):
-        self.diar_model = SortformerEncLabelModel.from_pretrained(model_name)
+        """
-        self.diar_model.eval()
+        Stores the shared streaming Sortformer diarization model. Used when a new online_diarization is initialized.
+        """
+        self._load_model(model_name)
+    
+    def _load_model(self, model_name: str):
+        """Load and configure the Sortformer model for streaming."""
+        try:
+            self.diar_model = SortformerEncLabelModel.from_pretrained(model_name)
+            self.diar_model.eval()
 
-        if torch.cuda.is_available():
+            device = torch.device("cuda" if torch.cuda.is_available() else "cpu")
-            self.diar_model.to(torch.device("cuda"))
+            self.diar_model.to(device)
+            
+            ## to test
+            # for name, param in self.diar_model.named_parameters():
+            #     if param.device != device:
+            #         raise RuntimeError(f"Parameter {name} is on {param.device} but should be on {device}")
+            
+            logger.info(f"Using {device.type.upper()} for Sortformer model")
 
-        # Streaming parameters for speed
+            self.diar_model.sortformer_modules.chunk_len = 10
-        self.diar_model.sortformer_modules.chunk_len = 12
+            self.diar_model.sortformer_modules.subsampling_factor = 10
-        self.diar_model.sortformer_modules.chunk_right_context = 1
+            self.diar_model.sortformer_modules.chunk_right_context = 0
-        self.diar_model.sortformer_modules.spkcache_len = 188
+            self.diar_model.sortformer_modules.chunk_left_context = 10
-        self.diar_model.sortformer_modules.fifo_len = 188
+            self.diar_model.sortformer_modules.spkcache_len = 188
-        self.diar_model.sortformer_modules.spkcache_update_period = 144
+            self.diar_model.sortformer_modules.fifo_len = 188
-        self.diar_model.sortformer_modules.log = False
+            self.diar_model.sortformer_modules.spkcache_update_period = 144
-        self.diar_model.sortformer_modules._check_streaming_parameters()
+            self.diar_model.sortformer_modules.log = False
-
+            self.diar_model.sortformer_modules._check_streaming_parameters()
-        self.batch_size = 1
+                        
-        self.processed_signal_offset = torch.zeros((self.batch_size,), dtype=torch.long, device=self.diar_model.device)
+        except Exception as e:
+            logger.error(f"Failed to load Sortformer model: {e}")
+            raise
+ 
+class SortformerDiarizationOnline:
+    def __init__(self, shared_model, sample_rate: int = 16000):
+        """
+        Initialize the streaming Sortformer diarization system.
         
-        self.audio_buffer = np.array([], dtype=np.float32)
+        Args:
-        self.sample_rate = 16000
+            sample_rate: Audio sample rate (default: 16000)
+            model_name: Pre-trained model name (default: "nvidia/diar_streaming_sortformer_4spk-v2")
+        """
+        self.sample_rate = sample_rate
         self.speaker_segments = []
-
+        self.buffer_audio = np.array([], dtype=np.float32)
-        self.streaming_state = self.diar_model.sortformer_modules.init_streaming_state(
+        self.segment_lock = threading.Lock()
-            batch_size=self.batch_size,
+        self.global_time_offset = 0.0
-            async_streaming=True,
+        self.processed_time = 0.0
-            device=self.diar_model.device
+        self.debug = False
+                
+        self.diar_model = shared_model.diar_model
+             
+        self.audio2mel = AudioToMelSpectrogramPreprocessor(
+            window_size=0.025,
+            normalize="NA",
+            n_fft=512,
+            features=128,
+            pad_to=0
         )
-        self.total_preds = torch.zeros((self.batch_size, 0, self.diar_model.sortformer_modules.n_spk), device=self.diar_model.device)
+        self.audio2mel.to(self.diar_model.device)
-
+        
-
+        self.chunk_duration_seconds = (
-    def _prepare_audio_signal(self, signal):
+            self.diar_model.sortformer_modules.chunk_len * 
-        audio_signal = torch.tensor(signal).unsqueeze(0).to(self.diar_model.device)
+            self.diar_model.sortformer_modules.subsampling_factor * 
-        audio_signal_length = torch.tensor([audio_signal.shape[1]]).to(self.diar_model.device)
+            self.diar_model.preprocessor._cfg.window_stride
-        processed_signal, processed_signal_length = self.diar_model.preprocessor(input_signal=audio_signal, length=audio_signal_length)
-        return processed_signal, processed_signal_length
-
-    def _create_streaming_loader(self, processed_signal, processed_signal_length):
-        streaming_loader = self.diar_model.sortformer_modules.streaming_feat_loader(
-            feat_seq=processed_signal,
-            feat_seq_length=processed_signal_length,
-            feat_seq_offset=self.processed_signal_offset,
         )
-        return streaming_loader
+        
+        self._init_streaming_state()
+        
+        self._previous_chunk_features = None
+        self._chunk_index = 0
+        self._len_prediction = None
+        
+        # Audio buffer to store PCM chunks for debugging
+        self.audio_buffer = []
+        
+        # Buffer for accumulating audio chunks until reaching chunk_duration_seconds
+        self.audio_chunk_buffer = []
+        self.accumulated_duration = 0.0
+        
+        logger.info("SortformerDiarization initialized successfully")
+
+
+    def _init_streaming_state(self):
+        """Initialize the streaming state for the model."""
+        batch_size = 1
+        device = self.diar_model.device
+        
+        self.streaming_state = StreamingSortformerState()
+        self.streaming_state.spkcache = torch.zeros(
+            (batch_size, self.diar_model.sortformer_modules.spkcache_len, self.diar_model.sortformer_modules.fc_d_model), 
+            device=device
+        )
+        self.streaming_state.spkcache_preds = torch.zeros(
+            (batch_size, self.diar_model.sortformer_modules.spkcache_len, self.diar_model.sortformer_modules.n_spk), 
+            device=device
+        )
+        self.streaming_state.spkcache_lengths = torch.zeros((batch_size,), dtype=torch.long, device=device)
+        self.streaming_state.fifo = torch.zeros(
+            (batch_size, self.diar_model.sortformer_modules.fifo_len, self.diar_model.sortformer_modules.fc_d_model), 
+            device=device
+        )
+        self.streaming_state.fifo_lengths = torch.zeros((batch_size,), dtype=torch.long, device=device)
+        self.streaming_state.mean_sil_emb = torch.zeros((batch_size, self.diar_model.sortformer_modules.fc_d_model), device=device)
+        self.streaming_state.n_sil_frames = torch.zeros((batch_size,), dtype=torch.long, device=device)
+        
+        # Initialize total predictions tensor
+        self.total_preds = torch.zeros((batch_size, 0, self.diar_model.sortformer_modules.n_spk), device=device)
+
+    def insert_silence(self, silence_duration: float):
+        """
+        Insert silence period by adjusting the global time offset.
+        
+        Args:
+            silence_duration: Duration of silence in seconds
+        """
+        with self.segment_lock:
+            self.global_time_offset += silence_duration
+        logger.debug(f"Inserted silence of {silence_duration:.2f}s, new offset: {self.global_time_offset:.2f}s")
 
     async def diarize(self, pcm_array: np.ndarray):
         """
-        Process an incoming audio chunk for diarization.
+        Process audio data for diarization in streaming fashion.
+        
+        Args:
+            pcm_array: Audio data as numpy array
         """
-        self.audio_buffer = np.concatenate([self.audio_buffer, pcm_array])
+        try:
-        
+            if self.debug:
-        # Process in fixed-size chunks (e.g., 1 second)
+                self.audio_buffer.append(pcm_array.copy())
-        chunk_size = self.sample_rate # 1 second of audio
-        
-        while len(self.audio_buffer) >= chunk_size:
-            chunk_to_process = self.audio_buffer[:chunk_size]
-            self.audio_buffer = self.audio_buffer[chunk_size:]
 
-            processed_signal, processed_signal_length = self._prepare_audio_signal(chunk_to_process)
+            threshold = int(self.chunk_duration_seconds * self.sample_rate)
             
-            current_offset_seconds = self.processed_signal_offset.item() * self.diar_model.preprocessor._cfg.window_stride
+            self.buffer_audio = np.concatenate([self.buffer_audio, pcm_array.copy()])
-
+            if not len(self.buffer_audio) >= threshold:
-            streaming_loader = self._create_streaming_loader(processed_signal, processed_signal_length)
+                return
             
-            frame_duration_s = self.diar_model.sortformer_modules.subsampling_factor * self.diar_model.preprocessor._cfg.window_stride
+            audio = self.buffer_audio[:threshold]
-            chunk_duration_seconds = self.diar_model.sortformer_modules.chunk_len * frame_duration_s
+            self.buffer_audio = self.buffer_audio[threshold:]
+            
+            device = self.diar_model.device
+            audio_signal_chunk = torch.tensor(audio, device=device).unsqueeze(0)
+            audio_signal_length_chunk = torch.tensor([audio_signal_chunk.shape[1]], device=device)
+            
+            processed_signal_chunk, processed_signal_length_chunk = self.audio2mel.get_features(
+                audio_signal_chunk, audio_signal_length_chunk
+            )
+            processed_signal_chunk = processed_signal_chunk.to(device)
+            processed_signal_length_chunk = processed_signal_length_chunk.to(device)
+            
+            if self._previous_chunk_features is not None:
+                to_add = self._previous_chunk_features[:, :, -99:].to(device)
+                total_features = torch.concat([to_add, processed_signal_chunk], dim=2).to(device)
+            else:
+                total_features = processed_signal_chunk.to(device)
+            
+            self._previous_chunk_features = processed_signal_chunk.to(device)
+            
+            chunk_feat_seq_t = torch.transpose(total_features, 1, 2).to(device)
+            
+            with torch.inference_mode():
+                left_offset = 8 if self._chunk_index > 0 else 0
+                right_offset = 8
+                
+                self.streaming_state, self.total_preds = self.diar_model.forward_streaming_step(
+                    processed_signal=chunk_feat_seq_t,
+                    processed_signal_length=torch.tensor([chunk_feat_seq_t.shape[1]]).to(device),
+                    streaming_state=self.streaming_state,
+                    total_preds=self.total_preds,
+                    left_offset=left_offset,
+                    right_offset=right_offset,
+                )
+                
+            # Convert predictions to speaker segments
+            self._process_predictions()
+            
+            self._chunk_index += 1
+            
+        except Exception as e:
+            logger.error(f"Error in diarize: {e}")
+            raise
+            
+        # TODO: Handle case when stream ends with partial buffer (accumulated_duration > 0 but < chunk_duration_seconds)
 
-            for i, chunk_feat_seq_t, feat_lengths, left_offset, right_offset in streaming_loader:
+    def _process_predictions(self):
-                with torch.inference_mode():
+        """Process model predictions and convert to speaker segments."""
-                    self.streaming_state, self.total_preds = self.diar_model.forward_streaming_step(
+        try:
-                        processed_signal=chunk_feat_seq_t,
+            preds_np = self.total_preds[0].cpu().numpy()
-                        processed_signal_length=feat_lengths,
+            active_speakers = np.argmax(preds_np, axis=1)
-                        streaming_state=self.streaming_state,
+            
-                        total_preds=self.total_preds,
+            if self._len_prediction is None:
-                        left_offset=left_offset,
+                self._len_prediction = len(active_speakers)
-                        right_offset=right_offset,
+            
-                    )
+            # Get predictions for current chunk
+            frame_duration = self.chunk_duration_seconds / self._len_prediction
+            current_chunk_preds = active_speakers[-self._len_prediction:]
+            
+            with self.segment_lock:
+                # Process predictions into segments
+                base_time = self._chunk_index * self.chunk_duration_seconds + self.global_time_offset
+                
+                for idx, spk in enumerate(current_chunk_preds):
+                    start_time = base_time + idx * frame_duration
+                    end_time = base_time + (idx + 1) * frame_duration
                     
-                    num_new_frames = feat_lengths[0].item()
+                    # Check if this continues the last segment or starts a new one
-                    
+                    if (self.speaker_segments and 
-                    # Get predictions for the current chunk from the end of total_preds
+                        self.speaker_segments[-1].speaker == spk and 
-                    preds_np = self.total_preds[0, -num_new_frames:].cpu().numpy()
+                        abs(self.speaker_segments[-1].end - start_time) < frame_duration * 0.5):
-                    active_speakers = np.argmax(preds_np, axis=1)
+                        # Continue existing segment
-
+                        self.speaker_segments[-1].end = end_time
-                    for idx, spk in enumerate(active_speakers):
+                    else:
-                        start_time = current_offset_seconds + (i * chunk_duration_seconds) + (idx * frame_duration_s)
-                        end_time = start_time + frame_duration_s
                         
-                        if self.speaker_segments and self.speaker_segments[-1].speaker == spk + 1:
+                        # Create new segment
-                            self.speaker_segments[-1].end = end_time
+                        self.speaker_segments.append(SpeakerSegment(
-                        else:
+                            speaker=spk,
-                            self.speaker_segments.append(SpeakerSegment(
+                            start=start_time,
-                                speaker=int(spk + 1),
+                            end=end_time
-                                start=start_time,
+                        ))
-                                end=end_time
+                
-                            ))
+                # Update processed time
-            
+                self.processed_time = max(self.processed_time, base_time + self.chunk_duration_seconds)
-            self.processed_signal_offset += processed_signal_length
+                
+                logger.debug(f"Processed chunk {self._chunk_index}, total segments: {len(self.speaker_segments)}")
+                
+        except Exception as e:
+            logger.error(f"Error processing predictions: {e}")
 
-
+    def assign_speakers_to_tokens(self, tokens: list, use_punctuation_split: bool = False) -> list:
-    def assign_speakers_to_tokens(self, tokens: list, **kwargs) -> list:
         """
         Assign speakers to tokens based on timing overlap with speaker segments.
+        
+        Args:
+            tokens: List of tokens with timing information
+            use_punctuation_split: Whether to use punctuation for boundary refinement
+            
+        Returns:
+            List of tokens with speaker assignments
         """
-        for token in tokens:
+        with self.segment_lock:
-            for segment in self.speaker_segments:
+            segments = self.speaker_segments.copy()
-                if not (segment.end <= token.start or segment.start >= token.end):
+        
-                    token.speaker = segment.speaker
+        if not segments or not tokens:
+            logger.debug("No segments or tokens available for speaker assignment")
+            return tokens
+        
+        logger.debug(f"Assigning speakers to {len(tokens)} tokens using {len(segments)} segments")
+        use_punctuation_split = False
+        if not use_punctuation_split:
+            # Simple overlap-based assignment
+            for token in tokens:
+                token.speaker = -1  # Default to no speaker
+                for segment in segments:
+                    # Check for timing overlap
+                    if not (segment.end <= token.start or segment.start >= token.end):
+                        token.speaker = segment.speaker + 1  # Convert to 1-based indexing
+                        break
+        else:
+            # Use punctuation-aware assignment (similar to diart_backend)
+            tokens = self._add_speaker_to_tokens_with_punctuation(segments, tokens)
+        
         return tokens
 
+    def _add_speaker_to_tokens_with_punctuation(self, segments: List[SpeakerSegment], tokens: list) -> list:
+        """
+        Assign speakers to tokens with punctuation-aware boundary adjustment.
+        
+        Args:
+            segments: List of speaker segments
+            tokens: List of tokens to assign speakers to
+            
+        Returns:
+            List of tokens with speaker assignments
+        """
+        punctuation_marks = {'.', '!', '?'}
+        punctuation_tokens = [token for token in tokens if token.text.strip() in punctuation_marks]
+        
+        # Convert segments to concatenated format
+        segments_concatenated = self._concatenate_speakers(segments)
+        
+        # Adjust segment boundaries based on punctuation
+        for ind, segment in enumerate(segments_concatenated):
+            for i, punctuation_token in enumerate(punctuation_tokens):
+                if punctuation_token.start > segment['end']:
+                    after_length = punctuation_token.start - segment['end']
+                    before_length = segment['end'] - punctuation_tokens[i - 1].end if i > 0 else float('inf')
+                    
+                    if before_length > after_length:
+                        segment['end'] = punctuation_token.start
+                        if i < len(punctuation_tokens) - 1 and ind + 1 < len(segments_concatenated):
+                            segments_concatenated[ind + 1]['begin'] = punctuation_token.start
+                    else:
+                        segment['end'] = punctuation_tokens[i - 1].end if i > 0 else segment['end']
+                        if i < len(punctuation_tokens) - 1 and ind - 1 >= 0:
+                            segments_concatenated[ind - 1]['begin'] = punctuation_tokens[i - 1].end
+                    break
+        
+        # Ensure non-overlapping tokens
+        last_end = 0.0
+        for token in tokens:
+            start = max(last_end + 0.01, token.start)
+            token.start = start
+            token.end = max(start, token.end)
+            last_end = token.end
+        
+        # Assign speakers based on adjusted segments
+        ind_last_speaker = 0
+        for segment in segments_concatenated:
+            for i, token in enumerate(tokens[ind_last_speaker:]):
+                if token.end <= segment['end']:
+                    token.speaker = segment['speaker']
+                    ind_last_speaker = i + 1
+                elif token.start > segment['end']:
+                    break
+        
+        return tokens
+
+    def _concatenate_speakers(self, segments: List[SpeakerSegment]) -> List[dict]:
+        """
+        Concatenate consecutive segments from the same speaker.
+        
+        Args:
+            segments: List of speaker segments
+            
+        Returns:
+            List of concatenated speaker segments
+        """
+        if not segments:
+            return []
+            
+        segments_concatenated = [{"speaker": segments[0].speaker + 1, "begin": segments[0].start, "end": segments[0].end}]
+        
+        for segment in segments[1:]:
+            speaker = segment.speaker + 1
+            if segments_concatenated[-1]['speaker'] != speaker:
+                segments_concatenated.append({"speaker": speaker, "begin": segment.start, "end": segment.end})
+            else:
+                segments_concatenated[-1]['end'] = segment.end
+                
+        return segments_concatenated
+
+    def get_segments(self) -> List[SpeakerSegment]:
+        """Get a copy of the current speaker segments."""
+        with self.segment_lock:
+            return self.speaker_segments.copy()
+
+    def clear_old_segments(self, older_than: float = 30.0):
+        """Clear segments older than the specified time."""
+        with self.segment_lock:
+            current_time = self.processed_time
+            self.speaker_segments = [
+                segment for segment in self.speaker_segments 
+                if current_time - segment.end < older_than
+            ]
+            logger.debug(f"Cleared old segments, remaining: {len(self.speaker_segments)}")
+
     def close(self):
-        """
+        """Close the diarization system and clean up resources."""
-        Cleanup resources.
+        logger.info("Closing SortformerDiarization")
-        """
+        with self.segment_lock:
-        logger.info("Closing SortformerDiarization.")
+            self.speaker_segments.clear()
+        
+        if self.debug:
+            concatenated_audio = np.concatenate(self.audio_buffer)
+            audio_data_int16 = (concatenated_audio * 32767).astype(np.int16)                
+            with wave.open("diarization_audio.wav", "wb") as wav_file:
+                wav_file.setnchannels(1)  # mono audio
+                wav_file.setsampwidth(2)   # 2 bytes per sample (int16)
+                wav_file.setframerate(self.sample_rate)
+                wav_file.writeframes(audio_data_int16.tobytes())
+            logger.info(f"Saved {len(concatenated_audio)} samples to diarization_audio.wav")
+
+
+def extract_number(s: str) -> int:
+    """Extract number from speaker string (compatibility function)."""
+    import re
+    m = re.search(r'\d+', s)
+    return int(m.group()) if m else 0
+
 
 if __name__ == '__main__':
+    import asyncio
     import librosa
-    an4_audio = 'new_audio_test.mp3'
+    
-    signal, sr = librosa.load(an4_audio, sr=16000)
+    async def main():
+        """TEST ONLY."""
+        an4_audio = 'audio_test.mp3'
+        signal, sr = librosa.load(an4_audio, sr=16000)
+        signal = signal[:16000*30]
 
-    diarization_pipeline = SortformerDiarization()
+        print("\n" + "=" * 50)
-
+        print("ground truth:")
-    # Simulate streaming
+        print("Speaker 0: 0:00 - 0:09")
-    chunk_size = 16000  # 1 second
+        print("Speaker 1: 0:09 - 0:19") 
-    for i in range(0, len(signal), chunk_size):
+        print("Speaker 2: 0:19 - 0:25")
-        chunk = signal[i:i+chunk_size]
+        print("Speaker 0: 0:25 - 0:30")
-        import asyncio
+        print("=" * 50)
-        asyncio.run(diarization_pipeline.diarize(chunk))
+        
-
+        diarization = SortformerDiarization(sample_rate=16000)        
-    for segment in diarization_pipeline.speaker_segments:
+        chunk_size = 1600
-        print(f"Speaker {segment.speaker}: {segment.start:.2f}s - {segment.end:.2f}s")
+        
+        for i in range(0, len(signal), chunk_size):
+            chunk = signal[i:i+chunk_size]
+            await diarization.diarize(chunk)
+            print(f"Processed chunk {i // chunk_size + 1}")
+        
+        segments = diarization.get_segments()
+        print("\nDiarization results:")
+        for segment in segments:
+            print(f"Speaker {segment.speaker}: {segment.start:.2f}s - {segment.end:.2f}s")
+    
+    asyncio.run(main())

whisperlivekit/diarization/sortformer_backend_2.py

@@ -1,257 +0,0 @@
-import numpy as np
-import torch
-import logging
-import math
-logger = logging.getLogger(__name__)
-
-try:
-    from nemo.collections.asr.models import SortformerEncLabelModel
-except ImportError:
-    raise SystemExit("""Please use `pip install "git+https://github.com/NVIDIA/NeMo.git@main#egg=nemo_toolkit[asr]"` to use the Sortformer diarization""")
-    
-
-diar_model = SortformerEncLabelModel.from_pretrained("nvidia/diar_streaming_sortformer_4spk-v2")
-diar_model.eval()
-
-if torch.cuda.is_available():
-    diar_model.to(torch.device("cuda"))
-    
-# Set the streaming parameters corresponding to 1.04s latency setup. This will affect the streaming feat loader.
-# diar_model.sortformer_modules.chunk_len = 6
-# diar_model.sortformer_modules.spkcache_len = 188
-# diar_model.sortformer_modules.chunk_right_context = 7
-# diar_model.sortformer_modules.fifo_len = 188
-# diar_model.sortformer_modules.spkcache_update_period = 144
-# diar_model.sortformer_modules.log = False
-
-
-# here we change the settings for our goal: speed!
-# we want batches of around 1 second. one frame is 0.08s, so 1s is 12.5 frames. we take 12.
-diar_model.sortformer_modules.chunk_len = 12
-
-# for more speed, we reduce the 'right context'. it's like looking less into the future.
-diar_model.sortformer_modules.chunk_right_context = 1
-
-# we keep the rest same for now
-diar_model.sortformer_modules.spkcache_len = 188
-diar_model.sortformer_modules.fifo_len = 188
-diar_model.sortformer_modules.spkcache_update_period = 144
-diar_model.sortformer_modules.log = False
-diar_model.sortformer_modules._check_streaming_parameters()
-
-batch_size = 1
-processed_signal_offset = torch.zeros((batch_size,), dtype=torch.long, device=diar_model.device)
-
-# from nemo.collections.asr.parts.preprocessing.features import FilterbankFeatures
-# from nemo.collections.asr.modules.audio_preprocessing import get_features
-from nemo.collections.asr.modules.audio_preprocessing import AudioToMelSpectrogramPreprocessor
-
-
-def prepare_audio_signal(signal):
-    audio_signal = torch.tensor(signal).unsqueeze(0).to(diar_model.device)
-    audio_signal_length = torch.tensor([audio_signal.shape[1]]).to(diar_model.device)
-    processed_signal, processed_signal_length = AudioToMelSpectrogramPreprocessor(
-            window_size= 0.025, 
-            normalize="NA",
-            n_fft=512,
-            features=128).get_features(audio_signal, audio_signal_length)
-    return processed_signal, processed_signal_length
-
-
-def streaming_feat_loader(
-    feat_seq, feat_seq_length, feat_seq_offset
-):
-    """
-    Load a chunk of feature sequence for streaming inference.
-
-    Args:
-        feat_seq (torch.Tensor): Tensor containing feature sequence
-            Shape: (batch_size, feat_dim, feat frame count)
-        feat_seq_length (torch.Tensor): Tensor containing feature sequence lengths
-            Shape: (batch_size,)
-        feat_seq_offset (torch.Tensor): Tensor containing feature sequence offsets
-            Shape: (batch_size,)
-
-    Returns:
-        chunk_idx (int): Index of the current chunk
-        chunk_feat_seq (torch.Tensor): Tensor containing the chunk of feature sequence
-            Shape: (batch_size, diar frame count, feat_dim)
-        feat_lengths (torch.Tensor): Tensor containing lengths of the chunk of feature sequence
-            Shape: (batch_size,)
-    """
-    feat_len = feat_seq.shape[2]
-    num_chunks = math.ceil(feat_len / (diar_model.sortformer_modules.chunk_len * diar_model.sortformer_modules.subsampling_factor))
-    if False:
-        logging.info(
-            f"feat_len={feat_len}, num_chunks={num_chunks}, "
-            f"feat_seq_length={feat_seq_length}, feat_seq_offset={feat_seq_offset}"
-        )
-
-    stt_feat, end_feat, chunk_idx = 0, 0, 0
-    while end_feat < feat_len:
-        left_offset = min(diar_model.sortformer_modules.chunk_left_context * diar_model.sortformer_modules.subsampling_factor, stt_feat)
-        end_feat = min(stt_feat + diar_model.sortformer_modules.chunk_len * diar_model.sortformer_modules.subsampling_factor, feat_len)
-        right_offset = min(diar_model.sortformer_modules.chunk_right_context * diar_model.sortformer_modules.subsampling_factor, feat_len - end_feat)
-        chunk_feat_seq = feat_seq[:, :, stt_feat - left_offset : end_feat + right_offset]
-        feat_lengths = (feat_seq_length + feat_seq_offset - stt_feat + left_offset).clamp(
-            0, chunk_feat_seq.shape[2]
-        )
-        feat_lengths = feat_lengths * (feat_seq_offset < end_feat)
-        stt_feat = end_feat
-        chunk_feat_seq_t = torch.transpose(chunk_feat_seq, 1, 2)
-        if False:
-            logging.info(
-                f"chunk_idx: {chunk_idx}, "
-                f"chunk_feat_seq_t shape: {chunk_feat_seq_t.shape}, "
-                f"chunk_feat_lengths: {feat_lengths}"
-            )
-        yield chunk_idx, chunk_feat_seq_t, feat_lengths, left_offset, right_offset
-        chunk_idx += 1
-
-
-class StreamingSortformerState:
-    """
-    This class creates a class instance that will be used to store the state of the
-    streaming Sortformer model.
-
-    Attributes:
-        spkcache (torch.Tensor): Speaker cache to store embeddings from start
-        spkcache_lengths (torch.Tensor): Lengths of the speaker cache
-        spkcache_preds (torch.Tensor): The speaker predictions for the speaker cache parts
-        fifo (torch.Tensor): FIFO queue to save the embedding from the latest chunks
-        fifo_lengths (torch.Tensor): Lengths of the FIFO queue
-        fifo_preds (torch.Tensor): The speaker predictions for the FIFO queue parts
-        spk_perm (torch.Tensor): Speaker permutation information for the speaker cache
-        mean_sil_emb (torch.Tensor): Mean silence embedding
-        n_sil_frames (torch.Tensor): Number of silence frames
-    """
-
-    spkcache = None  # Speaker cache to store embeddings from start
-    spkcache_lengths = None  #
-    spkcache_preds = None  # speaker cache predictions
-    fifo = None  # to save the embedding from the latest chunks
-    fifo_lengths = None
-    fifo_preds = None
-    spk_perm = None
-    mean_sil_emb = None
-    n_sil_frames = None
-
-
-def init_streaming_state(self, batch_size: int = 1, async_streaming: bool = False, device: torch.device = None):
-    """
-    Initializes StreamingSortformerState with empty tensors or zero-valued tensors.
-
-    Args:
-        batch_size (int): Batch size for tensors in streaming state
-        async_streaming (bool): True for asynchronous update, False for synchronous update
-        device (torch.device): Device for tensors in streaming state
-
-    Returns:
-        streaming_state (SortformerStreamingState): initialized streaming state
-    """
-    streaming_state = StreamingSortformerState()
-    if async_streaming:
-        streaming_state.spkcache = torch.zeros((batch_size, self.spkcache_len, self.fc_d_model), device=device)
-        streaming_state.spkcache_preds = torch.zeros((batch_size, self.spkcache_len, self.n_spk), device=device)
-        streaming_state.spkcache_lengths = torch.zeros((batch_size,), dtype=torch.long, device=device)
-        streaming_state.fifo = torch.zeros((batch_size, self.fifo_len, self.fc_d_model), device=device)
-        streaming_state.fifo_lengths = torch.zeros((batch_size,), dtype=torch.long, device=device)
-    else:
-        streaming_state.spkcache = torch.zeros((batch_size, 0, self.fc_d_model), device=device)
-        streaming_state.fifo = torch.zeros((batch_size, 0, self.fc_d_model), device=device)
-    streaming_state.mean_sil_emb = torch.zeros((batch_size, self.fc_d_model), device=device)
-    streaming_state.n_sil_frames = torch.zeros((batch_size,), dtype=torch.long, device=device)
-    return streaming_state
-
-def process_diarization(signal, chunks):
-    
-    audio_signal = torch.tensor(signal).unsqueeze(0).to(diar_model.device)
-    audio_signal_length = torch.tensor([audio_signal.shape[1]]).to(diar_model.device)
-    processed_signal, processed_signal_length = AudioToMelSpectrogramPreprocessor(
-            window_size= 0.025, 
-            normalize="NA",
-            n_fft=512,
-            features=128).get_features(audio_signal, audio_signal_length)
-
-    
-    streaming_loader = streaming_feat_loader(processed_signal, processed_signal_length, processed_signal_offset)
-
-    
-    streaming_state = init_streaming_state(diar_model.sortformer_modules,
-        batch_size = batch_size,
-        async_streaming = True,
-        device = diar_model.device
-    )
-    total_preds = torch.zeros((batch_size, 0, diar_model.sortformer_modules.n_spk), device=diar_model.device)
-
-    
-    chunk_duration_seconds = diar_model.sortformer_modules.chunk_len * diar_model.sortformer_modules.subsampling_factor * diar_model.preprocessor._cfg.window_stride
-    print(f"Chunk duration: {chunk_duration_seconds} seconds")
-
-    l_speakers = [
-        {'start_time': 0,
-        'end_time': 0,
-        'speaker': 0
-        }
-    ]
-    len_prediction = None
-    left_offset = 0
-    right_offset = 8
-    for i, chunk_feat_seq_t, _, _, _ in streaming_loader:
-        with torch.inference_mode():
-                streaming_state, total_preds = diar_model.forward_streaming_step(
-                    processed_signal=chunk_feat_seq_t,
-                    processed_signal_length=torch.tensor([chunk_feat_seq_t.shape[1]]),
-                    streaming_state=streaming_state,
-                    total_preds=total_preds,
-                    left_offset=left_offset,
-                    right_offset=right_offset,
-                )
-                left_offset = 8
-                preds_np = total_preds[0].cpu().numpy()
-                active_speakers = np.argmax(preds_np, axis=1)
-                if len_prediction is None:
-                    len_prediction = len(active_speakers) # we want to get the len of 1 prediction
-                frame_duration = chunk_duration_seconds / len_prediction
-                active_speakers = active_speakers[-len_prediction:]
-                print(chunk_feat_seq_t.shape, total_preds.shape)
-                for idx, spk in enumerate(active_speakers):
-                    if spk != l_speakers[-1]['speaker']:
-                        l_speakers.append(
-                            {'start_time': i * chunk_duration_seconds + idx * frame_duration,
-                            'end_time': i * chunk_duration_seconds + (idx + 1) * frame_duration,
-                            'speaker': spk
-                        })                    
-                    else:
-                        l_speakers[-1]['end_time'] = i * chunk_duration_seconds + (idx + 1) * frame_duration
-                    
-        print(l_speakers)
-        """
-        Should print
-        [{'start_time': 0, 'end_time': 8.72, 'speaker': 0}, 
-        {'start_time': 8.72, 'end_time': 18.88, 'speaker': 1},
-        {'start_time': 18.88, 'end_time': 24.96, 'speaker': 2},
-        {'start_time': 24.96, 'end_time': 31.68, 'speaker': 0}]
-        """
-
-if __name__ == '__main__':
-    import librosa
-    an4_audio = 'new_audio_test.mp3'
-    signal, sr = librosa.load(an4_audio,sr=16000) 
-
-    """
-    ground truth:
-    speaker 0 : 0:00 - 0:09
-    speaker 1 : 0:09 - 0:19
-    speaker 2 : 0:19 - 0:25
-    speaker 0 : 0:25 - end
-    """
-
-    # Simulate streaming
-    chunk_size = 16000  # 1 second
-    chunks = []
-    for i in range(0, len(signal), chunk_size):
-        chunk = signal[i:i+chunk_size]
-        chunks.append(chunk)
-
-    process_diarization(signal, chunks)

whisperlivekit/diarization/sortformer_backend_offline.py

@@ -0,0 +1,205 @@
+import numpy as np
+import torch
+import logging
+
+from nemo.collections.asr.models import SortformerEncLabelModel
+from nemo.collections.asr.modules import AudioToMelSpectrogramPreprocessor
+import librosa
+
+logger = logging.getLogger(__name__)
+
+def load_model():
+
+    diar_model = SortformerEncLabelModel.from_pretrained("nvidia/diar_streaming_sortformer_4spk-v2")
+    diar_model.eval()
+
+    if torch.cuda.is_available():
+        diar_model.to(torch.device("cuda"))
+
+    #we target 1 second lag for the moment. chunk_len could be reduced.
+    diar_model.sortformer_modules.chunk_len = 10
+    diar_model.sortformer_modules.subsampling_factor = 10 #8 would be better ideally
+
+    diar_model.sortformer_modules.chunk_right_context = 0 #no.
+    diar_model.sortformer_modules.chunk_left_context = 10 #big so it compensiate the problem with no padding later.
+
+    diar_model.sortformer_modules.spkcache_len = 188
+    diar_model.sortformer_modules.fifo_len = 188
+    diar_model.sortformer_modules.spkcache_update_period = 144
+    diar_model.sortformer_modules.log = False
+    diar_model.sortformer_modules._check_streaming_parameters()
+
+
+    audio2mel = AudioToMelSpectrogramPreprocessor(
+            window_size= 0.025, 
+            normalize="NA",
+            n_fft=512,
+            features=128,
+            pad_to=0) #pad_to 16 works better than 0. On test audio, we detect a third speaker for 1 second with pad_to=0. To solve that : increase left context to 10.
+
+    return diar_model, audio2mel
+
+diar_model, audio2mel = load_model()
+
+class StreamingSortformerState:
+    """
+    This class creates a class instance that will be used to store the state of the
+    streaming Sortformer model.
+
+    Attributes:
+        spkcache (torch.Tensor): Speaker cache to store embeddings from start
+        spkcache_lengths (torch.Tensor): Lengths of the speaker cache
+        spkcache_preds (torch.Tensor): The speaker predictions for the speaker cache parts
+        fifo (torch.Tensor): FIFO queue to save the embedding from the latest chunks
+        fifo_lengths (torch.Tensor): Lengths of the FIFO queue
+        fifo_preds (torch.Tensor): The speaker predictions for the FIFO queue parts
+        spk_perm (torch.Tensor): Speaker permutation information for the speaker cache
+        mean_sil_emb (torch.Tensor): Mean silence embedding
+        n_sil_frames (torch.Tensor): Number of silence frames
+    """
+
+    spkcache = None  # Speaker cache to store embeddings from start
+    spkcache_lengths = None  #
+    spkcache_preds = None  # speaker cache predictions
+    fifo = None  # to save the embedding from the latest chunks
+    fifo_lengths = None
+    fifo_preds = None
+    spk_perm = None
+    mean_sil_emb = None
+    n_sil_frames = None
+
+
+def init_streaming_state(self, batch_size: int = 1, async_streaming: bool = False, device: torch.device = None):
+    """
+    Initializes StreamingSortformerState with empty tensors or zero-valued tensors.
+
+    Args:
+        batch_size (int): Batch size for tensors in streaming state
+        async_streaming (bool): True for asynchronous update, False for synchronous update
+        device (torch.device): Device for tensors in streaming state
+
+    Returns:
+        streaming_state (SortformerStreamingState): initialized streaming state
+    """
+    streaming_state = StreamingSortformerState()
+    if async_streaming:
+        streaming_state.spkcache = torch.zeros((batch_size, self.spkcache_len, self.fc_d_model), device=device)
+        streaming_state.spkcache_preds = torch.zeros((batch_size, self.spkcache_len, self.n_spk), device=device)
+        streaming_state.spkcache_lengths = torch.zeros((batch_size,), dtype=torch.long, device=device)
+        streaming_state.fifo = torch.zeros((batch_size, self.fifo_len, self.fc_d_model), device=device)
+        streaming_state.fifo_lengths = torch.zeros((batch_size,), dtype=torch.long, device=device)
+    else:
+        streaming_state.spkcache = torch.zeros((batch_size, 0, self.fc_d_model), device=device)
+        streaming_state.fifo = torch.zeros((batch_size, 0, self.fc_d_model), device=device)
+    streaming_state.mean_sil_emb = torch.zeros((batch_size, self.fc_d_model), device=device)
+    streaming_state.n_sil_frames = torch.zeros((batch_size,), dtype=torch.long, device=device)
+    return streaming_state
+
+
+def process_diarization(chunks):
+    """ 
+    what it does:
+    1. Preprocessing: Applies dithering and pre-emphasis (high-pass filter) if enabled
+    2. STFT: Computes the Short-Time Fourier Transform using:
+        - the window of window_size=0.025 --> size of a window : 400 samples
+        - the hop parameter : n_window_stride = 0.01 -> every 160 samples, a new window
+    3. Magnitude Calculation: Converts complex STFT output to magnitude spectrogram
+    4. Mel Conversion: Applies Mel filterbanks (128 filters in this case) to get Mel spectrogram
+    5. Logarithm: Takes the log of the Mel spectrogram (if `log=True`)
+    6. Normalization: Skips normalization since `normalize="NA"`
+    7. Padding: Pads the time dimension to a multiple of `pad_to` (default 16)    
+    """
+    previous_chunk = None
+    l_chunk_feat_seq_t = []
+    for chunk in chunks:
+        audio_signal_chunk = torch.tensor(chunk).unsqueeze(0).to(diar_model.device)
+        audio_signal_length_chunk = torch.tensor([audio_signal_chunk.shape[1]]).to(diar_model.device)
+        processed_signal_chunk, processed_signal_length_chunk = audio2mel.get_features(audio_signal_chunk, audio_signal_length_chunk)
+        if previous_chunk is not None:
+            to_add = previous_chunk[:, :, -99:]
+            total = torch.concat([to_add, processed_signal_chunk], dim=2)
+        else:
+            total = processed_signal_chunk
+        previous_chunk = processed_signal_chunk
+        l_chunk_feat_seq_t.append(torch.transpose(total, 1, 2))
+
+    batch_size = 1
+    streaming_state = init_streaming_state(diar_model.sortformer_modules,
+        batch_size = batch_size,
+        async_streaming = True,
+        device = diar_model.device
+    )
+    total_preds = torch.zeros((batch_size, 0, diar_model.sortformer_modules.n_spk), device=diar_model.device)
+
+    chunk_duration_seconds = diar_model.sortformer_modules.chunk_len * diar_model.sortformer_modules.subsampling_factor * diar_model.preprocessor._cfg.window_stride
+
+    l_speakers = [
+        {'start_time': 0,
+        'end_time': 0,
+        'speaker': 0
+        }
+    ]
+    len_prediction = None
+    left_offset = 0
+    right_offset = 8
+    for i, chunk_feat_seq_t in enumerate(l_chunk_feat_seq_t):
+        with torch.inference_mode():
+                streaming_state, total_preds = diar_model.forward_streaming_step(
+                    processed_signal=chunk_feat_seq_t,
+                    processed_signal_length=torch.tensor([chunk_feat_seq_t.shape[1]]),
+                    streaming_state=streaming_state,
+                    total_preds=total_preds,
+                    left_offset=left_offset,
+                    right_offset=right_offset,
+                )
+                left_offset = 8
+                preds_np = total_preds[0].cpu().numpy()
+                active_speakers = np.argmax(preds_np, axis=1)
+                if len_prediction is None:
+                    len_prediction = len(active_speakers) # we want to get the len of 1 prediction
+                frame_duration = chunk_duration_seconds / len_prediction
+                active_speakers = active_speakers[-len_prediction:]
+                for idx, spk in enumerate(active_speakers):
+                    if spk != l_speakers[-1]['speaker']:
+                        l_speakers.append(
+                            {'start_time': (i * chunk_duration_seconds + idx * frame_duration),
+                            'end_time': (i * chunk_duration_seconds + (idx + 1) * frame_duration),
+                            'speaker': spk
+                        })                    
+                    else:
+                        l_speakers[-1]['end_time'] = i * chunk_duration_seconds + (idx + 1) * frame_duration
+                    
+        
+        """
+        Should print
+        [{'start_time': 0, 'end_time': 8.72, 'speaker': 0}, 
+        {'start_time': 8.72, 'end_time': 18.88, 'speaker': 1},
+        {'start_time': 18.88, 'end_time': 24.96, 'speaker': 2},
+        {'start_time': 24.96, 'end_time': 31.68, 'speaker': 0}]
+        """
+    for speaker in l_speakers:
+        print(f"Speaker {speaker['speaker']}: {speaker['start_time']:.2f}s - {speaker['end_time']:.2f}s")    
+    
+
+if __name__ == '__main__':
+    
+    an4_audio = 'audio_test.mp3'
+    signal, sr = librosa.load(an4_audio, sr=16000)
+    signal = signal[:16000*30]
+    # signal = signal[:-(len(signal)%16000)]
+
+    print("\n" + "=" * 50)
+    print("Expected ground truth:")
+    print("Speaker 0: 0:00 - 0:09")
+    print("Speaker 1: 0:09 - 0:19") 
+    print("Speaker 2: 0:19 - 0:25")
+    print("Speaker 0: 0:25 - 0:30")
+    print("=" * 50)
+
+    chunk_size = 16000  # 1 second
+    chunks = []
+    for i in range(0, len(signal), chunk_size):
+        chunk = signal[i:i+chunk_size]
+        chunks.append(chunk)
+    
+    process_diarization(chunks)

whisperlivekit/parse_args.py

@@ -61,7 +61,7 @@ def parse_args():
     parser.add_argument(
         "--diarization-backend",
         type=str,
-        default="diart",
+        default="sortformer",
         choices=["sortformer", "diart"],
         help="The diarization backend to use.",
     )
@@ -72,6 +72,12 @@ def parse_args():
         help="Disable transcription to only see live diarization results.",
     )
     
+    parser.add_argument(
+        "--disable-punctuation-split",
+        action="store_true",
+        help="Disable the split parameter.",
+    )
+    
     parser.add_argument(
         "--min-chunk-size",
         type=float,
@@ -112,6 +118,15 @@ def parse_args():
         choices=["transcribe", "translate"],
         help="Transcribe or translate.",
     )
+    
+    parser.add_argument(
+        "--target-language",
+        type=str,
+        default="",
+        dest="target_language",
+        help="Target language for translation. Not functional yet.",
+    )    
+
     parser.add_argument(
         "--backend",
         type=str,
@@ -158,9 +173,22 @@ def parse_args():
     )
     parser.add_argument("--ssl-certfile", type=str, help="Path to the SSL certificate file.", default=None)
     parser.add_argument("--ssl-keyfile", type=str, help="Path to the SSL private key file.", default=None)
-
+    parser.add_argument(
+        "--pcm-input",
+        action="store_true",
+        default=False,
+        help="If set, raw PCM (s16le) data is expected as input and FFmpeg will be bypassed."
+    )
     # SimulStreaming-specific arguments
     simulstreaming_group = parser.add_argument_group('SimulStreaming arguments (only used with --backend simulstreaming)')
+
+    simulstreaming_group.add_argument(
+        "--disable-fast-encoder",
+        action="store_true",
+        default=False,
+        dest="disable_fast_encoder",
+        help="Disable Faster Whisper or MLX Whisper backends for encoding (if installed). Slower but helpful when GPU memory is limited",
+    )
     
     simulstreaming_group.add_argument(
         "--frame-threshold",
@@ -252,10 +280,10 @@ def parse_args():
     )
     
     simulstreaming_group.add_argument(
-        "--preloaded_model_count",
+        "--preload-model-count",
         type=int,
         default=1,
-        dest="preloaded_model_count",
+        dest="preload_model_count",
         help="Optional. Number of models to preload in memory to speed up loading (set up to the expected number of concurrent instances).",
     )
 

whisperlivekit/remove_silences.py

@@ -81,7 +81,7 @@ def ends_with_silence(tokens, buffer_transcription, buffer_diarization, current_
     if not tokens:
         return [], buffer_transcription, buffer_diarization
     last_token = tokens[-1]
-    if tokens and (
+    if tokens and current_time and (
         current_time - last_token.end >= END_SILENCE_DURATION 
         or 
         (current_time - last_token.end >= 3 and vac_detected_silence)

whisperlivekit/results_formater.py

@@ -0,0 +1,137 @@
+
+import logging
+from whisperlivekit.remove_silences import handle_silences
+from whisperlivekit.timed_objects import Line, format_time
+
+logger = logging.getLogger(__name__)
+logger.setLevel(logging.DEBUG)
+
+PUNCTUATION_MARKS = {'.', '!', '?', '。', '！', '？'}
+CHECK_AROUND = 4
+
+def is_punctuation(token):
+    if token.text.strip() in PUNCTUATION_MARKS:
+        return True
+    return False
+
+def next_punctuation_change(i, tokens):
+    for ind in range(i+1, min(len(tokens), i+CHECK_AROUND+1)):
+        if is_punctuation(tokens[ind]):
+            return ind        
+    return None
+
+def next_speaker_change(i, tokens, speaker):
+    for ind in range(i-1, max(0, i-CHECK_AROUND)-1, -1):
+        token = tokens[ind]
+        if is_punctuation(token):
+            break
+        if token.speaker != speaker:
+            return ind, token.speaker
+    return None, speaker
+    
+def new_line(
+    token,
+    speaker,
+    debug_info = ""
+):
+    return Line(
+        speaker = speaker,
+        text = token.text + debug_info,
+        start = token.start,
+        end = token.end,
+    )
+
+def append_token_to_last_line(lines, sep, token, debug_info):
+    if token.text:
+        lines[-1].text += sep + token.text + debug_info
+        lines[-1].end = token.end
+
+def format_output(state, silence, current_time, args, debug):
+    diarization = args.diarization
+    disable_punctuation_split = args.disable_punctuation_split
+    tokens = state["tokens"]
+    translated_segments = state["translated_segments"] # Here we will attribute the speakers only based on the timestamps of the segments
+    buffer_transcription = state["buffer_transcription"]
+    buffer_diarization = state["buffer_diarization"]
+    end_attributed_speaker = state["end_attributed_speaker"]
+    sep = state["sep"]
+    
+    previous_speaker = -1
+    lines = []
+    undiarized_text = []
+    tokens, buffer_transcription, buffer_diarization = handle_silences(tokens, buffer_transcription, buffer_diarization, current_time, silence)
+    last_punctuation = None
+    for i, token in enumerate(tokens):
+        speaker = token.speaker
+        if not diarization and speaker == -1: #Speaker -1 means no attributed by diarization. In the frontend, it should appear under 'Speaker 1'
+            speaker = 1
+        if diarization and not tokens[-1].speaker == -2:
+            if (speaker in [-1, 0]) and token.end >= end_attributed_speaker:
+                undiarized_text.append(token.text)
+                continue
+            elif (speaker in [-1, 0]) and token.end < end_attributed_speaker:
+                speaker = previous_speaker
+        debug_info = ""
+        if debug:
+            debug_info = f"[{format_time(token.start)} : {format_time(token.end)}]"
+            
+        if not lines:
+            lines.append(new_line(token, speaker, debug_info = ""))
+            continue
+        else:
+            previous_speaker = lines[-1].speaker
+        
+        if is_punctuation(token):
+            last_punctuation = i
+            
+        
+        if last_punctuation == i-1:
+            if speaker != previous_speaker:
+                # perfect, diarization perfectly aligned
+                lines.append(new_line(token, speaker, debug_info = ""))
+                last_punctuation, next_punctuation = None, None
+                continue
+            
+            speaker_change_pos, new_speaker = next_speaker_change(i, tokens, speaker)
+            if speaker_change_pos:
+                # Corrects delay:
+                # That was the idea. Okay haha |SPLIT SPEAKER| that's a good one 
+                # should become:
+                # That was the idea. |SPLIT SPEAKER| Okay haha that's a good one 
+                lines.append(new_line(token, new_speaker, debug_info = ""))
+            else:
+                # No speaker change to come
+                append_token_to_last_line(lines, sep, token, debug_info)
+            continue
+        
+
+        if speaker != previous_speaker:
+            if speaker == -2 or previous_speaker == -2: #silences can happen anytime
+                lines.append(new_line(token, speaker, debug_info = ""))
+                continue
+            elif next_punctuation_change(i, tokens):
+                # Corrects advance:
+                # Are you |SPLIT SPEAKER| okay? yeah, sure. Absolutely 
+                # should become:
+                # Are you okay? |SPLIT SPEAKER| yeah, sure. Absolutely 
+                append_token_to_last_line(lines, sep, token, debug_info)
+                continue
+            else: #we create a new speaker, but that's no ideal. We are not sure about the split. We prefer to append to previous line
+                if disable_punctuation_split:
+                    lines.append(new_line(token, speaker, debug_info = ""))
+                    continue
+                pass
+            
+        append_token_to_last_line(lines, sep, token, debug_info)
+    if lines and translated_segments:
+        cts_idx = 0 # current_translated_segment_idx
+        for line in lines:
+            while cts_idx < len(translated_segments):
+                ts = translated_segments[cts_idx]
+                if ts.start and ts.start >= line.start and ts.end <= line.end:
+                    line.translation += ts.text + ' '
+                    cts_idx += 1
+                else:
+                    break
+    return lines, undiarized_text, buffer_transcription, '' 
+

whisperlivekit/simul_whisper/backend.py

@@ -3,25 +3,37 @@ import numpy as np
 import logging
 from typing import List, Tuple, Optional
 import logging
+import platform
 from whisperlivekit.timed_objects import ASRToken, Transcript
 from whisperlivekit.warmup import load_file
 from whisperlivekit.simul_whisper.license_simulstreaming import SIMULSTREAMING_LICENSE
 from .whisper import load_model, tokenizer
 from .whisper.audio import TOKENS_PER_SECOND
-
 import os
 import gc
 logger = logging.getLogger(__name__)
 
+import torch
+from whisperlivekit.simul_whisper.config import AlignAttConfig
+from whisperlivekit.simul_whisper.simul_whisper import PaddedAlignAttWhisper
+from whisperlivekit.simul_whisper.whisper import tokenizer
+
 try:
-    import torch
+    from .mlx_encoder import mlx_model_mapping, load_mlx_encoder
-    from whisperlivekit.simul_whisper.config import AlignAttConfig
+    HAS_MLX_WHISPER = True
-    from whisperlivekit.simul_whisper.simul_whisper import PaddedAlignAttWhisper
+except ImportError:
-    from whisperlivekit.simul_whisper.whisper import tokenizer
+    if platform.system() == "Darwin" and platform.machine() == "arm64":
-except ImportError as e:
+        print('MLX Whisper not found but you are on Apple Silicon. Consider installing mlx-whisper for better performance: pip install mlx-whisper')
-    raise ImportError(
+    HAS_MLX_WHISPER = False
-        """SimulStreaming dependencies are not available.
+if HAS_MLX_WHISPER:
-        Please install WhisperLiveKit using pip install "whisperlivekit[simulstreaming]".""")
+    HAS_FASTER_WHISPER = False
+else:
+    try:
+        from faster_whisper import WhisperModel
+        HAS_FASTER_WHISPER = True
+    except ImportError:
+        HAS_FASTER_WHISPER = False
+
 
 # TOO_MANY_REPETITIONS = 3
 
@@ -42,6 +54,8 @@ class SimulStreamingOnlineProcessor:
         self.committed: List[ASRToken] = []
         self.last_result_tokens: List[ASRToken] = []
         self.load_new_backend()
+        
+        #can be moved
         if asr.tokenizer:
             self.model.tokenizer = asr.tokenizer
 
@@ -49,7 +63,10 @@ class SimulStreamingOnlineProcessor:
         model = self.asr.get_new_model_instance()
         self.model = PaddedAlignAttWhisper(
             cfg=self.asr.cfg,
-            loaded_model=model)
+            loaded_model=model,
+            mlx_encoder=self.asr.mlx_encoder,
+            fw_encoder=self.asr.fw_encoder,
+            )
 
     def insert_silence(self, silence_duration, offset):
         """
@@ -62,7 +79,7 @@ class SimulStreamingOnlineProcessor:
         else:
             self.process_iter(is_last=True) #we want to totally process what remains in the buffer.
             self.model.refresh_segment(complete=True)
-            self.global_time_offset += silence_duration + offset
+            self.global_time_offset = silence_duration + offset
 
 
         
@@ -212,7 +229,7 @@ class SimulStreamingASR():
         logger.warning(SIMULSTREAMING_LICENSE)
         self.logfile = logfile
         self.transcribe_kargs = {}
-        self.original_language = None if lan == "auto" else lan
+        self.original_language = lan
         
         self.model_path = kwargs.get('model_path', './large-v3.pt')
         self.frame_threshold = kwargs.get('frame_threshold', 25)
@@ -229,7 +246,8 @@ class SimulStreamingASR():
         self.max_context_tokens = kwargs.get('max_context_tokens', None)
         self.warmup_file = kwargs.get('warmup_file', None)
         self.preload_model_count = kwargs.get('preload_model_count', 1)
-        
+        self.disable_fast_encoder = kwargs.get('disable_fast_encoder', False)
+        self.fast_encoder = False
         if model_dir is not None:
             self.model_path = model_dir
         elif modelsize is not None:
@@ -249,11 +267,6 @@ class SimulStreamingASR():
             }
             self.model_path = model_mapping.get(modelsize, f'./{modelsize}.pt')
         
-        # Set up tokenizer for translation if needed
-        if self.task == "translate":
-            self.tokenizer = self.set_translate_task()
-        else:
-            self.tokenizer = None
         self.cfg = AlignAttConfig(
                 model_path=self.model_path,
                 segment_length=self.segment_length,
@@ -271,17 +284,52 @@ class SimulStreamingASR():
                 static_init_prompt=self.static_init_prompt,
         )  
         
+        # Set up tokenizer for translation if needed
+        if self.task == "translate":
+            self.tokenizer = self.set_translate_task()
+        else:
+            self.tokenizer = None
+        
         self.model_name = os.path.basename(self.cfg.model_path).replace(".pt", "")
         self.model_path = os.path.dirname(os.path.abspath(self.cfg.model_path))
-        self.models = [self.load_model() for i in range(self.preload_model_count)]
     
+        self.mlx_encoder, self.fw_encoder = None, None
+        if not self.disable_fast_encoder:
+            if HAS_MLX_WHISPER:
+                print('Simulstreaming will use MLX whisper for a faster encoder.')
+                mlx_model_name = mlx_model_mapping[self.model_name]
+                self.mlx_encoder = load_mlx_encoder(path_or_hf_repo=mlx_model_name)
+                self.fast_encoder = True
+            elif HAS_FASTER_WHISPER:
+                print('Simulstreaming will use Faster Whisper for the encoder.')
+                self.fw_encoder = WhisperModel(
+                    self.model_name,
+                    device='auto',
+                    compute_type='auto',
+                )
+                self.fast_encoder = True
 
+        self.models = [self.load_model() for i in range(self.preload_model_count)]
 
 
     def load_model(self):
-        whisper_model = load_model(name=self.model_name, download_root=self.model_path)
+        whisper_model = load_model(name=self.model_name, download_root=self.model_path, decoder_only=self.fast_encoder)
         warmup_audio = load_file(self.warmup_file)
-        whisper_model.transcribe(warmup_audio, language=self.original_language)
+        if warmup_audio is not None:
+            warmup_audio = torch.from_numpy(warmup_audio).float()
+            if self.fast_encoder:                
+                temp_model = PaddedAlignAttWhisper(
+                    cfg=self.cfg,
+                    loaded_model=whisper_model,
+                    mlx_encoder=self.mlx_encoder,
+                    fw_encoder=self.fw_encoder,
+                )
+                temp_model.warmup(warmup_audio)
+                temp_model.remove_hooks()
+            else:
+                # For standard encoder, use the original transcribe warmup
+                warmup_audio = load_file(self.warmup_file)
+                whisper_model.transcribe(warmup_audio, language=self.original_language if self.original_language != 'auto' else None)
         return whisper_model
     
     def get_new_model_instance(self):
@@ -301,10 +349,12 @@ class SimulStreamingASR():
 
     def set_translate_task(self):
         """Set up translation task."""
+        if self.cfg.language == 'auto':
+            raise Exception('Translation cannot be done with language = auto')
         return tokenizer.get_tokenizer(
             multilingual=True,
-            language=self.model.cfg.language,
+            language=self.cfg.language,
-            num_languages=self.model.model.num_languages,
+            num_languages=99,
             task="translate"
         )
 

whisperlivekit/simul_whisper/mlx_encoder.py

@@ -0,0 +1,72 @@
+import json
+from pathlib import Path
+
+import mlx.core as mx
+import mlx.nn as nn
+from huggingface_hub import snapshot_download
+from mlx.utils import tree_unflatten
+
+from mlx_whisper import whisper
+
+mlx_model_mapping = {
+    "tiny.en": "mlx-community/whisper-tiny.en-mlx",
+    "tiny": "mlx-community/whisper-tiny-mlx",
+    "base.en": "mlx-community/whisper-base.en-mlx",
+    "base": "mlx-community/whisper-base-mlx",
+    "small.en": "mlx-community/whisper-small.en-mlx",
+    "small": "mlx-community/whisper-small-mlx",
+    "medium.en": "mlx-community/whisper-medium.en-mlx",
+    "medium": "mlx-community/whisper-medium-mlx",
+    "large-v1": "mlx-community/whisper-large-v1-mlx",
+    "large-v2": "mlx-community/whisper-large-v2-mlx",
+    "large-v3": "mlx-community/whisper-large-v3-mlx",
+    "large-v3-turbo": "mlx-community/whisper-large-v3-turbo",
+    "large": "mlx-community/whisper-large-mlx",
+}
+
+def load_mlx_encoder(
+    path_or_hf_repo: str,
+    dtype: mx.Dtype = mx.float32,
+) -> whisper.Whisper:
+    model_path = Path(path_or_hf_repo)
+    if not model_path.exists():
+        model_path = Path(snapshot_download(repo_id=path_or_hf_repo))
+
+    with open(str(model_path / "config.json"), "r") as f:
+        config = json.loads(f.read())
+        config.pop("model_type", None)
+        quantization = config.pop("quantization", None)
+
+    model_args = whisper.ModelDimensions(**config)
+
+    wf = model_path / "weights.safetensors"
+    if not wf.exists():
+        wf = model_path / "weights.npz"
+    weights = mx.load(str(wf))
+
+    model = whisper.Whisper(model_args, dtype)
+
+    if quantization is not None:
+        class_predicate = (
+            lambda p, m: isinstance(m, (nn.Linear, nn.Embedding))
+            and f"{p}.scales" in weights
+        )
+        nn.quantize(model, **quantization, class_predicate=class_predicate)
+
+    weights = tree_unflatten(list(weights.items()))
+    
+    # we only want to load the encoder weights here.
+    # Size examples: for tiny.en, 
+    # Decoder weights: 59110771 bytes
+    # Encoder weights: 15268874 bytes
+
+    
+    encoder_weights = {}
+    encoder_weights['encoder'] = weights['encoder']
+    del(weights)
+    
+
+
+    model.update(encoder_weights)
+    mx.eval(model.parameters())
+    return model

whisperlivekit/simul_whisper/simul_whisper.py

@@ -14,7 +14,7 @@ from .whisper.decoding import GreedyDecoder, BeamSearchDecoder, SuppressTokens,
 from .beam import BeamPyTorchInference
 from .eow_detection import fire_at_boundary, load_cif
 import os
-
+from time import time
 from .token_buffer import TokenBuffer
 
 import numpy as np
@@ -23,8 +23,22 @@ from .generation_progress import *
 DEC_PAD = 50257
 logger = logging.getLogger(__name__)
 
-import sys
+
-import wave
+try:
+    from mlx_whisper.audio import log_mel_spectrogram as mlx_log_mel_spectrogram
+    from mlx_whisper.transcribe import pad_or_trim as mlx_pad_or_trim
+    HAS_MLX_WHISPER = True
+except ImportError:
+    HAS_MLX_WHISPER = False
+if HAS_MLX_WHISPER:
+    HAS_FASTER_WHISPER = False
+else:
+    try:
+        from faster_whisper.audio import pad_or_trim as fw_pad_or_trim
+        from faster_whisper.feature_extractor import FeatureExtractor
+        HAS_FASTER_WHISPER = True
+    except ImportError:
+        HAS_FASTER_WHISPER = False
 
 # New features added to the original version of Simul-Whisper: 
 # - large-v3 model support
@@ -33,7 +47,13 @@ import wave
 # - prompt -- static vs. non-static
 # - context
 class PaddedAlignAttWhisper:
-    def __init__(self, cfg: AlignAttConfig, loaded_model=None) -> None:
+    def __init__(
+            self, 
+            cfg: AlignAttConfig,
+            loaded_model=None,
+            mlx_encoder=None,
+            fw_encoder=None,
+        ) -> None:
         self.log_segments = 0
         model_name = os.path.basename(cfg.model_path).replace(".pt", "")
         model_path = os.path.dirname(os.path.abspath(cfg.model_path))
@@ -41,7 +61,14 @@ class PaddedAlignAttWhisper:
             self.model = loaded_model
         else:
             self.model = load_model(name=model_name, download_root=model_path)
+        
+        self.device = 'cuda' if torch.cuda.is_available() else 'cpu'
 
+        self.mlx_encoder = mlx_encoder
+        self.fw_encoder = fw_encoder
+        if fw_encoder:
+            self.fw_feature_extractor = FeatureExtractor(feature_size=self.model.dims.n_mels)
+            
         logger.info(f"Model dimensions: {self.model.dims}")
 
         self.decode_options = DecodingOptions(
@@ -147,10 +174,18 @@ class PaddedAlignAttWhisper:
             self.token_decoder = BeamSearchDecoder(inference=self.inference, eot=self.tokenizer.eot, beam_size=cfg.beam_size)
             
     def remove_hooks(self):
-        print('remove hook')
         for hook in self.l_hooks:
             hook.remove()
 
+    def warmup(self, audio):
+        try:
+            self.insert_audio(audio)
+            self.infer(is_last=True)
+            self.refresh_segment(complete=True)
+            logger.info("Model warmed up successfully")
+        except Exception as e:
+            logger.exception(f"Model warmup failed: {e}")
+
     def create_tokenizer(self, language=None):
         self.tokenizer = tokenizer.get_tokenizer(
             multilingual=self.tokenizer_is_multilingual,  
@@ -359,20 +394,38 @@ class PaddedAlignAttWhisper:
         else:
             input_segments = self.segments[0]
 
-
+        # NEW : we can use a different encoder, before using standart whisper for cross attention with the hooks on the decoder
-        
+        beg_encode = time()
-        # mel + padding to 30s
+        if self.mlx_encoder:
-        mel_padded = log_mel_spectrogram(input_segments, n_mels=self.model.dims.n_mels, padding=N_SAMPLES, 
+            mlx_mel_padded = mlx_log_mel_spectrogram(audio=input_segments.detach(), n_mels=self.model.dims.n_mels, padding=N_SAMPLES)
-                                            device=self.model.device).unsqueeze(0)
+            mlx_mel = mlx_pad_or_trim(mlx_mel_padded, N_FRAMES, axis=-2)
-        # trim to 3000
+            mlx_encoder_feature = self.mlx_encoder.encoder(mlx_mel[None])
-        mel = pad_or_trim(mel_padded, N_FRAMES)
+            encoder_feature = torch.as_tensor(mlx_encoder_feature)
-
+            content_mel_len = int((mlx_mel_padded.shape[0] - mlx_mel.shape[0])/2)
-        # the len of actual audio
+        elif self.fw_encoder:
-        content_mel_len = int((mel_padded.shape[2] - mel.shape[2])/2)
+            audio_length_seconds = len(input_segments) / 16000   
-
+            content_mel_len = int(audio_length_seconds * 100)//2      
-        # encode
+            mel_padded_2 = self.fw_feature_extractor(waveform=input_segments.numpy(), padding=N_SAMPLES)[None, :]
-        encoder_feature = self.model.encoder(mel)
+            mel = fw_pad_or_trim(mel_padded_2, N_FRAMES, axis=-1)
-
+            encoder_feature_ctranslate = self.fw_encoder.encode(mel)
+            if self.device == 'cpu': #it seems that on gpu, passing StorageView to torch.as_tensor fails and wrapping in the array works
+                encoder_feature_ctranslate = np.array(encoder_feature_ctranslate)
+            try:
+                encoder_feature = torch.as_tensor(encoder_feature_ctranslate, device=self.device)
+            except TypeError: # Normally the cpu condition should prevent having exceptions, but just in case:
+                encoder_feature = torch.as_tensor(np.array(encoder_feature_ctranslate), device=self.device)
+        else:
+            # mel + padding to 30s
+            mel_padded = log_mel_spectrogram(input_segments, n_mels=self.model.dims.n_mels, padding=N_SAMPLES, 
+                                                device=self.device).unsqueeze(0)
+            # trim to 3000
+            mel = pad_or_trim(mel_padded, N_FRAMES)
+            # the len of actual audio
+            content_mel_len = int((mel_padded.shape[2] - mel.shape[2])/2)
+            encoder_feature = self.model.encoder(mel)
+        end_encode = time()
+        # print('Encoder duration:', end_encode-beg_encode)
+                
 #        logger.debug(f"Encoder feature shape: {encoder_feature.shape}")
 #        if mel.shape[-2:] != (self.model.dims.n_audio_ctx, self.model.dims.n_audio_state):
 #            logger.debug("mel ")
@@ -397,7 +450,7 @@ class PaddedAlignAttWhisper:
         ####################### Decoding loop
         logger.info("Decoding loop starts\n")
 
-        sum_logprobs = torch.zeros(self.cfg.beam_size, device=mel.device)
+        sum_logprobs = torch.zeros(self.cfg.beam_size, device=self.device)
         completed = False
 
         attn_of_alignment_heads = None
@@ -608,7 +661,7 @@ class PaddedAlignAttWhisper:
         ### new hypothesis
         logger.debug(f"new_hypothesis: {new_hypothesis}")
         new_tokens = torch.tensor([new_hypothesis], dtype=torch.long).repeat_interleave(self.cfg.beam_size, dim=0).to(
-            device=self.model.device,
+            device=self.device,
         )
         self.tokens.append(new_tokens)
         # TODO: test if this is redundant or not
@@ -618,4 +671,4 @@ class PaddedAlignAttWhisper:
         
         self._clean_cache()
 
-        return new_hypothesis, generation
+        return new_hypothesis, generation

whisperlivekit/simul_whisper/whisper/__init__.py

@@ -105,6 +105,7 @@ def load_model(
     device: Optional[Union[str, torch.device]] = None,
     download_root: str = None,
     in_memory: bool = False,
+    decoder_only=False
 ) -> Whisper:
     """
     Load a Whisper ASR model
@@ -151,7 +152,14 @@ def load_model(
     del checkpoint_file
 
     dims = ModelDimensions(**checkpoint["dims"])
-    model = Whisper(dims)
+    model = Whisper(dims, decoder_only=decoder_only)
+    
+    if decoder_only:
+        checkpoint["model_state_dict"] = {
+            k: v for k, v in checkpoint["model_state_dict"].items() 
+            if 'encoder' not in k
+        }
+
     model.load_state_dict(checkpoint["model_state_dict"])
 
     if alignment_heads is not None:

whisperlivekit/simul_whisper/whisper/model.py

@@ -253,16 +253,18 @@ class TextDecoder(nn.Module):
 
 
 class Whisper(nn.Module):
-    def __init__(self, dims: ModelDimensions):
+    def __init__(self, dims: ModelDimensions, decoder_only: bool = False):
         super().__init__()
         self.dims = dims
-        self.encoder = AudioEncoder(
+        
-            self.dims.n_mels,
+        if not decoder_only:
-            self.dims.n_audio_ctx,
+            self.encoder = AudioEncoder(
-            self.dims.n_audio_state,
+                self.dims.n_mels,
-            self.dims.n_audio_head,
+                self.dims.n_audio_ctx,
-            self.dims.n_audio_layer,
+                self.dims.n_audio_state,
-        )
+                self.dims.n_audio_head,
+                self.dims.n_audio_layer,
+            )
         self.decoder = TextDecoder(
             self.dims.n_vocab,
             self.dims.n_text_ctx,

whisperlivekit/timed_objects.py

@@ -1,10 +1,16 @@
 from dataclasses import dataclass
 from typing import Optional
+from datetime import timedelta
+
+def format_time(seconds: float) -> str:
+    """Format seconds as HH:MM:SS."""
+    return str(timedelta(seconds=int(seconds)))
+
 
 @dataclass
 class TimedText:
-    start: Optional[float]
+    start: Optional[float] = 0
-    end: Optional[float]
+    end: Optional[float] = 0
     text: Optional[str] = ''
     speaker: Optional[int] = -1
     probability: Optional[float] = None
@@ -31,6 +37,24 @@ class SpeakerSegment(TimedText):
     """
     pass
 
+@dataclass
+class Translation(TimedText):
+    pass
+
 @dataclass
 class Silence():
-    duration: float
+    duration: float
+    
+    
+@dataclass
+class Line(TimedText):
+    translation: str = ''
+    
+    def to_dict(self):
+        return {
+            'speaker': int(self.speaker),
+            'text': self.text,
+            'translation': self.translation,
+            'start': format_time(self.start),
+            'end': format_time(self.end),
+        }

whisperlivekit/translation/mapping_languages.py

@@ -0,0 +1,182 @@
+"""
+adapted from https://store.crowdin.com/custom-mt
+"""
+
+LANGUAGES = [
+    {"name": "Afrikaans", "nllb": "afr_Latn", "crowdin": "af"},
+    {"name": "Akan", "nllb": "aka_Latn", "crowdin": "ak"},
+    {"name": "Amharic", "nllb": "amh_Ethi", "crowdin": "am"},
+    {"name": "Assamese", "nllb": "asm_Beng", "crowdin": "as"},
+    {"name": "Asturian", "nllb": "ast_Latn", "crowdin": "ast"},
+    {"name": "Bashkir", "nllb": "bak_Cyrl", "crowdin": "ba"},
+    {"name": "Bambara", "nllb": "bam_Latn", "crowdin": "bm"},
+    {"name": "Balinese", "nllb": "ban_Latn", "crowdin": "ban"},
+    {"name": "Belarusian", "nllb": "bel_Cyrl", "crowdin": "be"},
+    {"name": "Bengali", "nllb": "ben_Beng", "crowdin": "bn"},
+    {"name": "Bosnian", "nllb": "bos_Latn", "crowdin": "bs"},
+    {"name": "Bulgarian", "nllb": "bul_Cyrl", "crowdin": "bg"},
+    {"name": "Catalan", "nllb": "cat_Latn", "crowdin": "ca"},
+    {"name": "Cebuano", "nllb": "ceb_Latn", "crowdin": "ceb"},
+    {"name": "Czech", "nllb": "ces_Latn", "crowdin": "cs"},
+    {"name": "Welsh", "nllb": "cym_Latn", "crowdin": "cy"},
+    {"name": "Danish", "nllb": "dan_Latn", "crowdin": "da"},
+    {"name": "German", "nllb": "deu_Latn", "crowdin": "de"},
+    {"name": "Dzongkha", "nllb": "dzo_Tibt", "crowdin": "dz"},
+    {"name": "Greek", "nllb": "ell_Grek", "crowdin": "el"},
+    {"name": "English", "nllb": "eng_Latn", "crowdin": "en"},
+    {"name": "Esperanto", "nllb": "epo_Latn", "crowdin": "eo"},
+    {"name": "Estonian", "nllb": "est_Latn", "crowdin": "et"},
+    {"name": "Basque", "nllb": "eus_Latn", "crowdin": "eu"},
+    {"name": "Ewe", "nllb": "ewe_Latn", "crowdin": "ee"},
+    {"name": "Faroese", "nllb": "fao_Latn", "crowdin": "fo"},
+    {"name": "Fijian", "nllb": "fij_Latn", "crowdin": "fj"},
+    {"name": "Finnish", "nllb": "fin_Latn", "crowdin": "fi"},
+    {"name": "French", "nllb": "fra_Latn", "crowdin": "fr"},
+    {"name": "Friulian", "nllb": "fur_Latn", "crowdin": "fur-IT"},
+    {"name": "Scottish Gaelic", "nllb": "gla_Latn", "crowdin": "gd"},
+    {"name": "Irish", "nllb": "gle_Latn", "crowdin": "ga-IE"},
+    {"name": "Galician", "nllb": "glg_Latn", "crowdin": "gl"},
+    {"name": "Guarani", "nllb": "grn_Latn", "crowdin": "gn"},
+    {"name": "Gujarati", "nllb": "guj_Gujr", "crowdin": "gu-IN"},
+    {"name": "Haitian Creole", "nllb": "hat_Latn", "crowdin": "ht"},
+    {"name": "Hausa", "nllb": "hau_Latn", "crowdin": "ha"},
+    {"name": "Hebrew", "nllb": "heb_Hebr", "crowdin": "he"},
+    {"name": "Hindi", "nllb": "hin_Deva", "crowdin": "hi"},
+    {"name": "Croatian", "nllb": "hrv_Latn", "crowdin": "hr"},
+    {"name": "Hungarian", "nllb": "hun_Latn", "crowdin": "hu"},
+    {"name": "Armenian", "nllb": "hye_Armn", "crowdin": "hy-AM"},
+    {"name": "Igbo", "nllb": "ibo_Latn", "crowdin": "ig"},
+    {"name": "Indonesian", "nllb": "ind_Latn", "crowdin": "id"},
+    {"name": "Icelandic", "nllb": "isl_Latn", "crowdin": "is"},
+    {"name": "Italian", "nllb": "ita_Latn", "crowdin": "it"},
+    {"name": "Javanese", "nllb": "jav_Latn", "crowdin": "jv"},
+    {"name": "Japanese", "nllb": "jpn_Jpan", "crowdin": "ja"},
+    {"name": "Kabyle", "nllb": "kab_Latn", "crowdin": "kab"},
+    {"name": "Kannada", "nllb": "kan_Knda", "crowdin": "kn"},
+    {"name": "Georgian", "nllb": "kat_Geor", "crowdin": "ka"},
+    {"name": "Kazakh", "nllb": "kaz_Cyrl", "crowdin": "kk"},
+    {"name": "Khmer", "nllb": "khm_Khmr", "crowdin": "km"},
+    {"name": "Kinyarwanda", "nllb": "kin_Latn", "crowdin": "rw"},
+    {"name": "Kyrgyz", "nllb": "kir_Cyrl", "crowdin": "ky"},
+    {"name": "Korean", "nllb": "kor_Hang", "crowdin": "ko"},
+    {"name": "Lao", "nllb": "lao_Laoo", "crowdin": "lo"},
+    {"name": "Ligurian", "nllb": "lij_Latn", "crowdin": "lij"},
+    {"name": "Limburgish", "nllb": "lim_Latn", "crowdin": "li"},
+    {"name": "Lingala", "nllb": "lin_Latn", "crowdin": "ln"},
+    {"name": "Lithuanian", "nllb": "lit_Latn", "crowdin": "lt"},
+    {"name": "Luxembourgish", "nllb": "ltz_Latn", "crowdin": "lb"},
+    {"name": "Maithili", "nllb": "mai_Deva", "crowdin": "mai"},
+    {"name": "Malayalam", "nllb": "mal_Mlym", "crowdin": "ml-IN"},
+    {"name": "Marathi", "nllb": "mar_Deva", "crowdin": "mr"},
+    {"name": "Macedonian", "nllb": "mkd_Cyrl", "crowdin": "mk"},
+    {"name": "Maltese", "nllb": "mlt_Latn", "crowdin": "mt"},
+    {"name": "Mossi", "nllb": "mos_Latn", "crowdin": "mos"},
+    {"name": "Maori", "nllb": "mri_Latn", "crowdin": "mi"},
+    {"name": "Burmese", "nllb": "mya_Mymr", "crowdin": "my"},
+    {"name": "Dutch", "nllb": "nld_Latn", "crowdin": "nl"},
+    {"name": "Norwegian Nynorsk", "nllb": "nno_Latn", "crowdin": "nn-NO"},
+    {"name": "Nepali", "nllb": "npi_Deva", "crowdin": "ne-NP"},
+    {"name": "Northern Sotho", "nllb": "nso_Latn", "crowdin": "nso"},
+    {"name": "Occitan", "nllb": "oci_Latn", "crowdin": "oc"},
+    {"name": "Odia", "nllb": "ory_Orya", "crowdin": "or"},
+    {"name": "Papiamento", "nllb": "pap_Latn", "crowdin": "pap"},
+    {"name": "Polish", "nllb": "pol_Latn", "crowdin": "pl"},
+    {"name": "Portuguese", "nllb": "por_Latn", "crowdin": "pt-PT"},
+    {"name": "Dari", "nllb": "prs_Arab", "crowdin": "fa-AF"},
+    {"name": "Romanian", "nllb": "ron_Latn", "crowdin": "ro"},
+    {"name": "Rundi", "nllb": "run_Latn", "crowdin": "rn"},
+    {"name": "Russian", "nllb": "rus_Cyrl", "crowdin": "ru"},
+    {"name": "Sango", "nllb": "sag_Latn", "crowdin": "sg"},
+    {"name": "Sanskrit", "nllb": "san_Deva", "crowdin": "sa"},
+    {"name": "Santali", "nllb": "sat_Olck", "crowdin": "sat"},
+    {"name": "Sinhala", "nllb": "sin_Sinh", "crowdin": "si-LK"},
+    {"name": "Slovak", "nllb": "slk_Latn", "crowdin": "sk"},
+    {"name": "Slovenian", "nllb": "slv_Latn", "crowdin": "sl"},
+    {"name": "Shona", "nllb": "sna_Latn", "crowdin": "sn"},
+    {"name": "Sindhi", "nllb": "snd_Arab", "crowdin": "sd"},
+    {"name": "Somali", "nllb": "som_Latn", "crowdin": "so"},
+    {"name": "Southern Sotho", "nllb": "sot_Latn", "crowdin": "st"},
+    {"name": "Spanish", "nllb": "spa_Latn", "crowdin": "es-ES"},
+    {"name": "Sardinian", "nllb": "srd_Latn", "crowdin": "sc"},
+    {"name": "Swati", "nllb": "ssw_Latn", "crowdin": "ss"},
+    {"name": "Sundanese", "nllb": "sun_Latn", "crowdin": "su"},
+    {"name": "Swedish", "nllb": "swe_Latn", "crowdin": "sv-SE"},
+    {"name": "Swahili", "nllb": "swh_Latn", "crowdin": "sw"},
+    {"name": "Tamil", "nllb": "tam_Taml", "crowdin": "ta"},
+    {"name": "Tatar", "nllb": "tat_Cyrl", "crowdin": "tt-RU"},
+    {"name": "Telugu", "nllb": "tel_Telu", "crowdin": "te"},
+    {"name": "Tajik", "nllb": "tgk_Cyrl", "crowdin": "tg"},
+    {"name": "Tagalog", "nllb": "tgl_Latn", "crowdin": "tl"},
+    {"name": "Thai", "nllb": "tha_Thai", "crowdin": "th"},
+    {"name": "Tigrinya", "nllb": "tir_Ethi", "crowdin": "ti"},
+    {"name": "Tswana", "nllb": "tsn_Latn", "crowdin": "tn"},
+    {"name": "Tsonga", "nllb": "tso_Latn", "crowdin": "ts"},
+    {"name": "Turkmen", "nllb": "tuk_Latn", "crowdin": "tk"},
+    {"name": "Turkish", "nllb": "tur_Latn", "crowdin": "tr"},
+    {"name": "Uyghur", "nllb": "uig_Arab", "crowdin": "ug"},
+    {"name": "Ukrainian", "nllb": "ukr_Cyrl", "crowdin": "uk"},
+    {"name": "Venetian", "nllb": "vec_Latn", "crowdin": "vec"},
+    {"name": "Vietnamese", "nllb": "vie_Latn", "crowdin": "vi"},
+    {"name": "Wolof", "nllb": "wol_Latn", "crowdin": "wo"},
+    {"name": "Xhosa", "nllb": "xho_Latn", "crowdin": "xh"},
+    {"name": "Yoruba", "nllb": "yor_Latn", "crowdin": "yo"},
+    {"name": "Zulu", "nllb": "zul_Latn", "crowdin": "zu"},
+]
+
+NAME_TO_NLLB = {lang["name"]: lang["nllb"] for lang in LANGUAGES}
+NAME_TO_CROWDIN = {lang["name"]: lang["crowdin"] for lang in LANGUAGES}
+CROWDIN_TO_NLLB = {lang["crowdin"]: lang["nllb"] for lang in LANGUAGES}
+NLLB_TO_CROWDIN = {lang["nllb"]: lang["crowdin"] for lang in LANGUAGES}
+CROWDIN_TO_NAME = {lang["crowdin"]: lang["name"] for lang in LANGUAGES}
+NLLB_TO_NAME = {lang["nllb"]: lang["name"] for lang in LANGUAGES}
+
+
+def get_nllb_code(crowdin_code):
+    return CROWDIN_TO_NLLB.get(crowdin_code, None)
+
+
+def get_crowdin_code(nllb_code):
+    return NLLB_TO_CROWDIN.get(nllb_code)
+
+
+def get_language_name_by_crowdin(crowdin_code):
+    return CROWDIN_TO_NAME.get(crowdin_code)
+
+
+def get_language_name_by_nllb(nllb_code):
+    return NLLB_TO_NAME.get(nllb_code)
+
+
+def get_language_info(identifier, identifier_type="auto"):
+    if identifier_type == "auto":
+        for lang in LANGUAGES:
+            if (lang["name"].lower() == identifier.lower() or 
+                lang["nllb"] == identifier or 
+                lang["crowdin"] == identifier):
+                return lang
+    elif identifier_type == "name":
+        for lang in LANGUAGES:
+            if lang["name"].lower() == identifier.lower():
+                return lang
+    elif identifier_type == "nllb":
+        for lang in LANGUAGES:
+            if lang["nllb"] == identifier:
+                return lang
+    elif identifier_type == "crowdin":
+        for lang in LANGUAGES:
+            if lang["crowdin"] == identifier:
+                return lang
+    
+    return None
+
+
+def list_all_languages():
+    return [lang["name"] for lang in LANGUAGES]
+
+
+def list_all_nllb_codes():
+    return [lang["nllb"] for lang in LANGUAGES]
+
+
+def list_all_crowdin_codes():
+    return [lang["crowdin"] for lang in LANGUAGES]

whisperlivekit/translation/translation.py

@@ -0,0 +1,137 @@
+import ctranslate2
+import torch
+import transformers
+from dataclasses import dataclass
+import huggingface_hub
+from whisperlivekit.translation.mapping_languages import get_nllb_code
+from whisperlivekit.timed_objects import Translation
+
+
+#In diarization case, we may want to translate just one speaker, or at least start the sentences there
+
+PUNCTUATION_MARKS = {'.', '!', '?', '。', '！', '？'}
+
+
+@dataclass
+class TranslationModel():
+    translator: ctranslate2.Translator
+    tokenizer: dict
+
+def load_model(src_langs):
+    MODEL = 'nllb-200-distilled-600M-ctranslate2'
+    MODEL_GUY = 'entai2965'
+    huggingface_hub.snapshot_download(MODEL_GUY + '/' + MODEL,local_dir=MODEL)
+    device = "cuda" if torch.cuda.is_available() else "cpu"
+    translator = ctranslate2.Translator(MODEL,device=device)
+    tokenizer = dict()
+    for src_lang in src_langs:
+        tokenizer[src_lang] = transformers.AutoTokenizer.from_pretrained(MODEL, src_lang=src_lang, clean_up_tokenization_spaces=True)
+    return TranslationModel(
+        translator=translator,
+        tokenizer=tokenizer
+    )
+
+def translate(input, translation_model, tgt_lang):
+    source = translation_model.tokenizer.convert_ids_to_tokens(translation_model.tokenizer.encode(input))
+    target_prefix = [tgt_lang]
+    results = translation_model.translator.translate_batch([source], target_prefix=[target_prefix])
+    target = results[0].hypotheses[0][1:]
+    return translation_model.tokenizer.decode(translation_model.tokenizer.convert_tokens_to_ids(target))
+
+class OnlineTranslation:
+    def __init__(self, translation_model: TranslationModel, input_languages: list, output_languages: list):
+        self.buffer = []
+        self.len_processed_buffer = 0
+        self.translation_remaining = Translation()
+        self.validated = []
+        self.translation_pending_validation = ''
+        self.translation_model = translation_model
+        self.input_languages = input_languages
+        self.output_languages = output_languages
+
+    def compute_common_prefix(self, results):
+        #we dont want want to prune the result for the moment. 
+        if not self.buffer:
+            self.buffer = results
+        else:
+            for i in range(min(len(self.buffer), len(results))):
+                if self.buffer[i] != results[i]:
+                    self.commited.extend(self.buffer[:i])
+                    self.buffer = results[i:]
+
+    def translate(self, input, input_lang=None, output_lang=None):
+        if not input:
+            return ""
+        if input_lang is None:
+            input_lang = self.input_languages[0]
+        if output_lang is None:
+            output_lang = self.output_languages[0]
+        nllb_output_lang = get_nllb_code(output_lang)
+            
+        source = self.translation_model.tokenizer[input_lang].convert_ids_to_tokens(self.translation_model.tokenizer[input_lang].encode(input))   
+        results = self.translation_model.translator.translate_batch([source], target_prefix=[[nllb_output_lang]]) #we can use return_attention=True to try to optimize the stuff.
+        target = results[0].hypotheses[0][1:]
+        results = self.translation_model.tokenizer[input_lang].decode(self.translation_model.tokenizer[input_lang].convert_tokens_to_ids(target))
+        return results
+
+    def translate_tokens(self, tokens):
+        if tokens:
+            text = ' '.join([token.text for token in tokens])
+            start = tokens[0].start
+            end = tokens[-1].end
+            translated_text = self.translate(text)
+            translation = Translation(
+                text=translated_text,
+                start=start,
+                end=end,
+            )
+            return translation
+        return None
+            
+        
+
+    def insert_tokens(self, tokens):
+        self.buffer.extend(tokens)
+        pass
+    
+    def process(self):
+        i = 0
+        if len(self.buffer) < self.len_processed_buffer + 3: #nothing new to process
+            return self.validated + [self.translation_remaining]
+        while i < len(self.buffer):
+            if self.buffer[i].text in PUNCTUATION_MARKS:
+                translation_sentence = self.translate_tokens(self.buffer[:i+1])
+                self.validated.append(translation_sentence)
+                self.buffer = self.buffer[i+1:]
+                i = 0
+            else:
+                i+=1
+        self.translation_remaining = self.translate_tokens(self.buffer)
+        self.len_processed_buffer = len(self.buffer)
+        return self.validated + [self.translation_remaining]
+                
+
+if __name__ == '__main__':
+    output_lang = 'fr'
+    input_lang = "en"
+    
+    
+    test_string = """
+    Transcription technology has improved so much in the past few years. Have you noticed how accurate real-time speech-to-text is now?
+    """
+    test = test_string.split(' ')
+    step = len(test) // 3
+    
+    shared_model = load_model([input_lang])
+    online_translation = OnlineTranslation(shared_model, input_languages=[input_lang], output_languages=[output_lang])
+        
+    for id in range(5):
+        val = test[id*step : (id+1)*step]
+        val_str = ' '.join(val)
+        result = online_translation.translate(val_str)
+        print(result)
+    
+    
+    
+    
+    # print(result)

whisperlivekit/warmup.py

@@ -31,21 +31,21 @@ def load_file(warmup_file=None, timeout=5):
                 logger.debug(f"Download successful in {time.time() - start_time:.2f}s")
             except (urllib.error.URLError, socket.timeout) as e:
                 logger.warning(f"Download failed: {e}. Proceeding without warmup.")
-                return False
+                return None
             finally:
                 socket.setdefaulttimeout(original_timeout)
     elif not warmup_file:
-        return False 
+        return None 
     
     if not warmup_file or not os.path.exists(warmup_file) or os.path.getsize(warmup_file) == 0:
         logger.warning(f"Warmup file {warmup_file} invalid or missing.")
-        return False
+        return None
     
     try:
         audio, sr = librosa.load(warmup_file, sr=16000)
     except Exception as e:
         logger.warning(f"Failed to load audio file: {e}")
-        return False
+        return None
     return audio
 
 def warmup_asr(asr, warmup_file=None, timeout=5):

whisperlivekit/web/live_transcription.css

@@ -74,10 +74,13 @@
 
 body {
   font-family: ui-sans-serif, system-ui, sans-serif, 'Apple Color Emoji', 'Segoe UI Emoji', 'Segoe UI Symbol', 'Noto Color Emoji';
-  margin: 20px;
+  margin: 0;
   text-align: center;
   background-color: var(--bg);
   color: var(--text);
+  height: 100vh;
+  display: flex;
+  flex-direction: column;
 }
 
 /* Record button */
@@ -168,9 +171,18 @@ body {
 }
 
 #status {
-  margin-top: 20px;
+  margin-top: 15px;
   font-size: 16px;
   color: var(--text);
+  margin-bottom: 0;
+}
+
+.header-container {
+  position: sticky;
+  top: 0;
+  background-color: var(--bg);
+  z-index: 100;
+  padding: 20px;
 }
 
 /* Settings */
@@ -179,12 +191,11 @@ body {
   justify-content: center;
   align-items: center;
   gap: 15px;
-  margin-top: 20px;
 }
 
 .settings {
   display: flex;
-  flex-direction: column;
+  flex-wrap: wrap;
   align-items: flex-start;
   gap: 12px;
 }
@@ -198,23 +209,27 @@ body {
 
 #chunkSelector,
 #websocketInput,
-#themeSelector {
+#themeSelector,
+#microphoneSelect {
   font-size: 16px;
   padding: 5px 8px;
   border-radius: 8px;
   border: 1px solid var(--border);
   background-color: var(--button-bg);
   color: var(--text);
-  max-height: 34px;
+  max-height: 30px;
 }
 
-#websocketInput {
+#microphoneSelect {
-  width: 220px;
+  width: 100%;
+  max-width: 190px;
+  min-width: 120px;
 }
 
 #chunkSelector:focus,
 #websocketInput:focus,
-#themeSelector:focus {
+#themeSelector:focus,
+#microphoneSelect:focus {
   outline: none;
   border-color: #007bff;
   box-shadow: 0 0 0 3px rgba(0, 123, 255, 0.15);
@@ -247,9 +262,9 @@ label {
 }
 
 .theme-selector-container {
-  position: absolute;
+  display: flex;
-  top: 20px;
+  align-items: center;
-  right: 20px;
+  margin-top: 17px;
 }
 
 .segmented label {
@@ -293,9 +308,21 @@ label {
   border-radius: 999px;
 }
 
+.transcript-container {
+  flex: 1;
+  overflow-y: auto;
+  padding: 20px;
+  scrollbar-width: none;
+  -ms-overflow-style: none;
+}
+
+.transcript-container::-webkit-scrollbar {
+  display: none;
+}
+
 /* Transcript area */
 #linesTranscript {
-  margin: 20px auto;
+  margin: 0 auto;
   max-width: 700px;
   text-align: left;
   font-size: 16px;
@@ -341,6 +368,27 @@ label {
   color: var(--label-trans-text);
 }
 
+.label_translation {
+  background-color: var(--chip-bg);
+  border-radius: 10px;
+  padding: 4px 8px;
+  margin-top: 4px;
+  font-size: 14px;
+  color: var(--text);
+  display: flex;
+  align-items: flex-start;
+  gap: 4px;
+}
+
+.label_translation img {
+  margin-top: 2px;
+}
+
+.label_translation img {
+  width: 12px;
+  height: 12px;
+}
+
 #timeInfo {
   color: var(--muted);
   margin-left: 10px;
@@ -390,6 +438,7 @@ label {
   font-size: 13px;
   border-radius: 30px;
   padding: 2px 10px;
+  display: none;
 }
 
 .loading {
@@ -400,3 +449,69 @@ label {
   font-size: 14px;
   margin-bottom: 0px;
 }
+
+/* for smaller screens */
+@media (max-width: 768px) {
+  .header-container {
+    padding: 15px;
+  }
+  
+  .settings-container {
+    flex-direction: column;
+    gap: 10px;
+  }
+  
+  .settings {
+    justify-content: center;
+    gap: 8px;
+  }
+  
+  .field {
+    align-items: center;
+  }
+  
+  #websocketInput,
+  #microphoneSelect {
+    min-width: 100px;
+    max-width: 160px;
+  }
+  
+  .theme-selector-container {
+    margin-top: 10px;
+  }
+  
+  .transcript-container {
+    padding: 15px;
+  }
+}
+
+@media (max-width: 480px) {
+  .header-container {
+    padding: 10px;
+  }
+  
+  .settings {
+    flex-direction: column;
+    align-items: center;
+    gap: 6px;
+  }
+  
+  #websocketInput,
+  #microphoneSelect {
+    max-width: 140px;
+  }
+  
+  .segmented label {
+    padding: 4px 8px;
+    font-size: 12px;
+  }
+  
+  .segmented img {
+    width: 14px;
+    height: 14px;
+  }
+  
+  .transcript-container {
+    padding: 10px;
+  }
+}

whisperlivekit/web/live_transcription.html

@@ -1,61 +1,73 @@
 <!DOCTYPE html>
 <html lang="en">
+
 <head>
-  <meta charset="UTF-8" />
+    <meta charset="UTF-8" />
-  <meta name="viewport" content="width=device-width, initial-scale=1.0" />
+    <meta name="viewport" content="width=device-width, initial-scale=1.0" />
-  <title>WhisperLiveKit</title>
+    <title>WhisperLiveKit</title>
-  <link rel="stylesheet" href="/web/live_transcription.css" />
+    <link rel="stylesheet" href="/web/live_transcription.css" />
 </head>
+
 <body>
-  <div class="settings-container">
+    <div class="header-container">
-    <button id="recordButton">
+        <div class="settings-container">
-      <div class="shape-container">
+            <button id="recordButton">
-        <div class="shape"></div>
+                <div class="shape-container">
-      </div>
+                    <div class="shape"></div>
-      <div class="recording-info">
+                </div>
-        <div class="wave-container">
+                <div class="recording-info">
-          <canvas id="waveCanvas"></canvas>
+                    <div class="wave-container">
+                        <canvas id="waveCanvas"></canvas>
+                    </div>
+                    <div class="timer">00:00</div>
+                </div>
+            </button>
+
+            <div class="settings">
+                <div class="field">
+                    <label for="websocketInput">Websocket URL</label>
+                    <input id="websocketInput" type="text" placeholder="ws://host:port/asr" />
+                </div>
+
+                <div class="field">
+                    <label id="microphoneSelectLabel" for="microphoneSelect">Select Microphone</label>
+                    <select id="microphoneSelect">
+                        <option value="">Default Microphone</option>
+                    </select>
+                </div>
+
+                <div class="theme-selector-container">
+                    <div class="segmented" role="radiogroup" aria-label="Theme selector">
+                        <input type="radio" id="theme-system" name="theme" value="system" />
+                        <label for="theme-system" title="System">
+                            <img src="/web/src/system_mode.svg" alt="" />
+                            <span>System</span>
+                        </label>
+
+                        <input type="radio" id="theme-light" name="theme" value="light" />
+                        <label for="theme-light" title="Light">
+                            <img src="/web/src/light_mode.svg" alt="" />
+                            <span>Light</span>
+                        </label>
+
+                        <input type="radio" id="theme-dark" name="theme" value="dark" />
+                        <label for="theme-dark" title="Dark">
+                            <img src="/web/src/dark_mode.svg" alt="" />
+                            <span>Dark</span>
+                        </label>
+                    </div>
+                </div>
+            </div>
         </div>
-        <div class="timer">00:00</div>
+        
-      </div>
+        <p id="status"></p>
-    </button>
-
-    <div class="settings">
-      <div class="field">
-        <label for="websocketInput">WebSocket URL</label>
-        <input id="websocketInput" type="text" placeholder="ws://host:port/asr" />
-      </div>
-
-      </div>
     </div>
-  </div>
 
-  <div class="theme-selector-container">
+    <div class="transcript-container">
-    <div class="segmented" role="radiogroup" aria-label="Theme selector">
+        <div id="linesTranscript"></div>
-      <input type="radio" id="theme-system" name="theme" value="system" />
-      <label for="theme-system" title="System">
-        <img src="/web/src/system_mode.svg" alt="" />
-        <span>System</span>
-      </label>
-
-      <input type="radio" id="theme-light" name="theme" value="light" />
-      <label for="theme-light" title="Light">
-        <img src="/web/src/light_mode.svg" alt="" />
-        <span>Light</span>
-      </label>
-
-      <input type="radio" id="theme-dark" name="theme" value="dark" />
-      <label for="theme-dark" title="Dark">
-        <img src="/web/src/dark_mode.svg" alt="" />
-        <span>Dark</span>
-      </label>
     </div>
-  </div>
 
-  <p id="status"></p>
+    <script src="/web/live_transcription.js"></script>
-
-  <div id="linesTranscript"></div>
-
-  <script src="/web/live_transcription.js"></script>
 </body>
-</html>
+
+</html>

whisperlivekit/web/live_transcription.js

@@ -18,6 +18,8 @@ let animationFrame = null;
 let waitingForStop = false;
 let lastReceivedData = null;
 let lastSignature = null;
+let availableMicrophones = [];
+let selectedMicrophoneId = null;
 
 waveCanvas.width = 60 * (window.devicePixelRatio || 1);
 waveCanvas.height = 30 * (window.devicePixelRatio || 1);
@@ -31,6 +33,7 @@ const websocketDefaultSpan = document.getElementById("wsDefaultUrl");
 const linesTranscriptDiv = document.getElementById("linesTranscript");
 const timerElement = document.querySelector(".timer");
 const themeRadios = document.querySelectorAll('input[name="theme"]');
+const microphoneSelect = document.getElementById("microphoneSelect");
 
 function getWaveStroke() {
   const styles = getComputedStyle(document.documentElement);
@@ -82,6 +85,61 @@ if (darkMq && darkMq.addEventListener) {
   darkMq.addListener(handleOsThemeChange);
 }
 
+async function enumerateMicrophones() {
+  try {
+    const stream = await navigator.mediaDevices.getUserMedia({ audio: true });
+    stream.getTracks().forEach(track => track.stop());
+
+    const devices = await navigator.mediaDevices.enumerateDevices();
+    availableMicrophones = devices.filter(device => device.kind === 'audioinput');
+
+    populateMicrophoneSelect();
+    console.log(`Found ${availableMicrophones.length} microphone(s)`);
+  } catch (error) {
+    console.error('Error enumerating microphones:', error);
+    statusText.textContent = "Error accessing microphones. Please grant permission.";
+  }
+}
+
+function populateMicrophoneSelect() {
+  if (!microphoneSelect) return;
+
+  microphoneSelect.innerHTML = '<option value="">Default Microphone</option>';
+
+  availableMicrophones.forEach((device, index) => {
+    const option = document.createElement('option');
+    option.value = device.deviceId;
+    option.textContent = device.label || `Microphone ${index + 1}`;
+    microphoneSelect.appendChild(option);
+  });
+
+  const savedMicId = localStorage.getItem('selectedMicrophone');
+  if (savedMicId && availableMicrophones.some(mic => mic.deviceId === savedMicId)) {
+    microphoneSelect.value = savedMicId;
+    selectedMicrophoneId = savedMicId;
+  }
+}
+
+function handleMicrophoneChange() {
+  selectedMicrophoneId = microphoneSelect.value || null;
+  localStorage.setItem('selectedMicrophone', selectedMicrophoneId || '');
+
+  const selectedDevice = availableMicrophones.find(mic => mic.deviceId === selectedMicrophoneId);
+  const deviceName = selectedDevice ? selectedDevice.label : 'Default Microphone';
+
+  console.log(`Selected microphone: ${deviceName}`);
+  statusText.textContent = `Microphone changed to: ${deviceName}`;
+
+  if (isRecording) {
+    statusText.textContent = "Switching microphone... Please wait.";
+    stopRecording().then(() => {
+      setTimeout(() => {
+        toggleRecording();
+      }, 1000);
+    });
+  }
+}
+
 // Helpers
 function fmt1(x) {
   const n = Number(x);
@@ -235,7 +293,7 @@ function renderLinesWithBuffer(
   const showTransLag = !isFinalizing && remaining_time_transcription > 0;
   const showDiaLag = !isFinalizing && !!buffer_diarization && remaining_time_diarization > 0;
   const signature = JSON.stringify({
-    lines: (lines || []).map((it) => ({ speaker: it.speaker, text: it.text, beg: it.beg, end: it.end })),
+    lines: (lines || []).map((it) => ({ speaker: it.speaker, text: it.text, start: it.start, end: it.end })),
     buffer_transcription: buffer_transcription || "",
     buffer_diarization: buffer_diarization || "",
     status: current_status,
@@ -258,8 +316,8 @@ function renderLinesWithBuffer(
   const linesHtml = (lines || [])
     .map((item, idx) => {
       let timeInfo = "";
-      if (item.beg !== undefined && item.end !== undefined) {
+      if (item.start !== undefined && item.end !== undefined) {
-        timeInfo = ` ${item.beg} - ${item.end}`;
+        timeInfo = ` ${item.start} - ${item.end}`;
       }
 
       let speakerLabel = "";
@@ -274,6 +332,13 @@ function renderLinesWithBuffer(
       }
 
       let currentLineText = item.text || "";
+      
+      if (item.translation) {
+        currentLineText += `<div class="label_translation">
+          <img src="/web/src/translate.svg" alt="Translation" width="12" height="12" />
+          <span>${item.translation}</span>
+        </div>`;
+      }
 
       if (idx === lines.length - 1) {
         if (!isFinalizing && item.speaker !== -2) {
@@ -315,7 +380,10 @@ function renderLinesWithBuffer(
     .join("");
 
   linesTranscriptDiv.innerHTML = linesHtml;
-  window.scrollTo({ top: document.body.scrollHeight, behavior: "smooth" });
+  const transcriptContainer = document.querySelector('.transcript-container');
+  if (transcriptContainer) {
+    transcriptContainer.scrollTo({ top: transcriptContainer.scrollHeight, behavior: "smooth" });
+  }
 }
 
 function updateTimer() {
@@ -377,7 +445,11 @@ async function startRecording() {
       console.log("Error acquiring wake lock.");
     }
 
-    const stream = await navigator.mediaDevices.getUserMedia({ audio: true });
+    const audioConstraints = selectedMicrophoneId 
+      ? { audio: { deviceId: { exact: selectedMicrophoneId } } }
+      : { audio: true };
+
+    const stream = await navigator.mediaDevices.getUserMedia(audioConstraints);
 
     audioContext = new (window.AudioContext || window.webkitAudioContext)();
     analyser = audioContext.createAnalyser();
@@ -400,7 +472,12 @@ async function startRecording() {
     isRecording = true;
     updateUI();
   } catch (err) {
-    statusText.textContent = "Error accessing microphone. Please allow microphone access.";
+    if (window.location.hostname === "0.0.0.0") {
+      statusText.textContent =
+        "Error accessing microphone. Browsers may block microphone access on 0.0.0.0. Try using localhost:8000 instead.";
+    } else {
+      statusText.textContent = "Error accessing microphone. Please allow microphone access.";
+    }
     console.error(err);
   }
 }
@@ -511,3 +588,22 @@ function updateUI() {
 }
 
 recordButton.addEventListener("click", toggleRecording);
+
+if (microphoneSelect) {
+  microphoneSelect.addEventListener("change", handleMicrophoneChange);
+}
+document.addEventListener('DOMContentLoaded', async () => {
+  try {
+    await enumerateMicrophones();
+  } catch (error) {
+    console.log("Could not enumerate microphones on load:", error);
+  }
+});
+navigator.mediaDevices.addEventListener('devicechange', async () => {
+  console.log('Device change detected, re-enumerating microphones');
+  try {
+    await enumerateMicrophones();
+  } catch (error) {
+    console.log("Error re-enumerating microphones:", error);
+  }
+});

whisperlivekit/web/src/settings.svg

@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" height="24px" viewBox="0 -960 960 960" width="24px" fill="#5f6368"><path d="M433-80q-27 0-46.5-18T363-142l-9-66q-13-5-24.5-12T307-235l-62 26q-25 11-50 2t-39-32l-47-82q-14-23-8-49t27-43l53-40q-1-7-1-13.5v-27q0-6.5 1-13.5l-53-40q-21-17-27-43t8-49l47-82q14-23 39-32t50 2l62 26q11-8 23-15t24-12l9-66q4-26 23.5-44t46.5-18h94q27 0 46.5 18t23.5 44l9 66q13 5 24.5 12t22.5 15l62-26q25-11 50-2t39 32l47 82q14 23 8 49t-27 43l-53 40q1 7 1 13.5v27q0 6.5-2 13.5l53 40q21 17 27 43t-8 49l-48 82q-14 23-39 32t-50-2l-60-26q-11 8-23 15t-24 12l-9 66q-4 26-23.5 44T527-80h-94Zm7-80h79l14-106q31-8 57.5-23.5T639-327l99 41 39-68-86-65q5-14 7-29.5t2-31.5q0-16-2-31.5t-7-29.5l86-65-39-68-99 42q-22-23-48.5-38.5T533-694l-13-106h-79l-14 106q-31 8-57.5 23.5T321-633l-99-41-39 68 86 64q-5 15-7 30t-2 32q0 16 2 31t7 30l-86 65 39 68 99-42q22 23 48.5 38.5T427-266l13 106Zm42-180q58 0 99-41t41-99q0-58-41-99t-99-41q-59 0-99.5 41T342-480q0 58 40.5 99t99.5 41Zm-2-140Z"/></svg>

whisperlivekit/web/src/translate.svg

@@ -0,0 +1 @@
+<svg xmlns="http://www.w3.org/2000/svg" height="24px" viewBox="0 -960 960 960" width="24px" fill="#5f6368"><path d="m603-202-34 97q-4 11-14 18t-22 7q-20 0-32.5-16.5T496-133l152-402q5-11 15-18t22-7h30q12 0 22 7t15 18l152 403q8 19-4 35.5T868-80q-13 0-22.5-7T831-106l-34-96H603ZM362-401 188-228q-11 11-27.5 11.5T132-228q-11-11-11-28t11-28l174-174q-35-35-63.5-80T190-640h84q20 39 40 68t48 58q33-33 68.5-92.5T484-720H80q-17 0-28.5-11.5T40-760q0-17 11.5-28.5T80-800h240v-40q0-17 11.5-28.5T360-880q17 0 28.5 11.5T400-840v40h240q17 0 28.5 11.5T680-760q0 17-11.5 28.5T640-720h-76q-21 72-63 148t-83 116l96 98-30 82-122-125Zm266 129h144l-72-204-72 204Z"/></svg>

whisperlivekit/web/web_interface.py

@@ -1,5 +1,6 @@
 import logging
 import importlib.resources as resources
+import base64
 
 logger = logging.getLogger(__name__)
 
@@ -12,6 +13,60 @@ def get_web_interface_html():
         logger.error(f"Error loading web interface HTML: {e}")
         return "<html><body><h1>Error loading interface</h1></body></html>"
 
+def get_inline_ui_html():
+    """Returns the complete web interface HTML with all assets embedded in a single call."""
+    try:
+        with resources.files('whisperlivekit.web').joinpath('live_transcription.html').open('r', encoding='utf-8') as f:
+            html_content = f.read()        
+        with resources.files('whisperlivekit.web').joinpath('live_transcription.css').open('r', encoding='utf-8') as f:
+            css_content = f.read()
+        with resources.files('whisperlivekit.web').joinpath('live_transcription.js').open('r', encoding='utf-8') as f:
+            js_content = f.read()
+        
+        # SVG files
+        with resources.files('whisperlivekit.web').joinpath('src', 'system_mode.svg').open('r', encoding='utf-8') as f:
+            system_svg = f.read()
+            system_data_uri = f"data:image/svg+xml;base64,{base64.b64encode(system_svg.encode('utf-8')).decode('utf-8')}"
+        with resources.files('whisperlivekit.web').joinpath('src', 'light_mode.svg').open('r', encoding='utf-8') as f:
+            light_svg = f.read()
+            light_data_uri = f"data:image/svg+xml;base64,{base64.b64encode(light_svg.encode('utf-8')).decode('utf-8')}"
+        with resources.files('whisperlivekit.web').joinpath('src', 'dark_mode.svg').open('r', encoding='utf-8') as f:
+            dark_svg = f.read()
+            dark_data_uri = f"data:image/svg+xml;base64,{base64.b64encode(dark_svg.encode('utf-8')).decode('utf-8')}"
+        
+        # Replace external references
+        html_content = html_content.replace(
+            '<link rel="stylesheet" href="/web/live_transcription.css" />',
+            f'<style>\n{css_content}\n</style>'
+        )
+        
+        html_content = html_content.replace(
+            '<script src="/web/live_transcription.js"></script>',
+            f'<script>\n{js_content}\n</script>'
+        )
+        
+        # Replace SVG references
+        html_content = html_content.replace(
+            '<img src="/web/src/system_mode.svg" alt="" />',
+            f'<img src="{system_data_uri}" alt="" />'
+        )
+        
+        html_content = html_content.replace(
+            '<img src="/web/src/light_mode.svg" alt="" />',
+            f'<img src="{light_data_uri}" alt="" />'
+        )
+        
+        html_content = html_content.replace(
+            '<img src="/web/src/dark_mode.svg" alt="" />',
+            f'<img src="{dark_data_uri}" alt="" />'
+        )
+        
+        return html_content
+        
+    except Exception as e:
+        logger.error(f"Error creating embedded web interface: {e}")
+        return "<html><body><h1>Error loading embedded interface</h1></body></html>"
+
 
 if __name__ == '__main__':
     
@@ -28,6 +83,6 @@ if __name__ == '__main__':
     
     @app.get("/")
     async def get():
-        return HTMLResponse(get_web_interface_html())
+        return HTMLResponse(get_inline_ui_html())
 
-    uvicorn.run(app=app)
+    uvicorn.run(app=app)