Update README.md

script to lib

README.md

@@ -5,7 +5,7 @@
 This project is based on [Whisper Streaming](https://github.com/ufal/whisper_streaming) and lets you transcribe audio directly from your browser. Simply launch the local server and grant microphone access. Everything runs locally on your machine ✨
 
 <p align="center">
-  <img src="https://raw.githubusercontent.com/QuentinFuxa/WhisperLiveKit/demo.png" alt="Demo Screenshot" width="730">
+  <img src="https://raw.githubusercontent.com/QuentinFuxa/WhisperLiveKit/refs/heads/main/demo.png" alt="Demo Screenshot" width="730">
 </p>
 
 ### Differences from [Whisper Streaming](https://github.com/ufal/whisper_streaming)
@@ -26,7 +26,7 @@ This project is based on [Whisper Streaming](https://github.com/ufal/whisper_str
 
 ## Installation
 
-### Via pip
+### Via pip (recommended)
 
 ```bash
 pip install whisperlivekit
@@ -46,88 +46,103 @@ pip install whisperlivekit
 
 You need to install FFmpeg on your system:
 
-- Install system dependencies:
+```bash
-    ```bash
+# For Ubuntu/Debian:
-    # Install FFmpeg on your system (required for audio processing)
+sudo apt install ffmpeg
-    # For Ubuntu/Debian:
-    sudo apt install ffmpeg
-    
-    # For macOS:
-    brew install ffmpeg
-    
-    # For Windows:
-    # Download from https://ffmpeg.org/download.html and add to PATH
-    ```
 
-- Install required Python dependencies:
+# For macOS:
+brew install ffmpeg
 
-    ```bash
+# For Windows:
-    # Whisper streaming required dependencies
+# Download from https://ffmpeg.org/download.html and add to PATH
-    pip install librosa soundfile
+```
 
-    # Whisper streaming web required dependencies
+### Optional Dependencies
-    pip install fastapi ffmpeg-python
-    ```
-- Install at least one whisper backend among:
 
-    ```
+```bash
-   whisper
+# If you want to use VAC (Voice Activity Controller). Useful for preventing hallucinations
-   whisper-timestamped
+pip install torch
-   faster-whisper (faster backend on NVIDIA GPU)
-   mlx-whisper (faster backend on Apple Silicon)
-   ```
-- Optionnal dependencies
-
-    ```
-    # If you want to use VAC (Voice Activity Controller). Useful for preventing hallucinations
-    torch
    
-    # If you choose sentences as buffer trimming strategy
+# If you choose sentences as buffer trimming strategy
-    mosestokenizer
+pip install mosestokenizer wtpsplit
-    wtpsplit
+pip install tokenize_uk  # If you work with Ukrainian text
-    tokenize_uk # If you work with Ukrainian text
 
-    # If you want to run the server using uvicorn (recommended)
+# If you want to use diarization
-    uvicorn
+pip install diart
+```
 
-    # If you want to use diarization
+Diart uses [pyannote.audio](https://github.com/pyannote/pyannote-audio) models from the _huggingface hub_. To use them, please follow the steps described [here](https://github.com/juanmc2005/diart?tab=readme-ov-file#get-access-to--pyannote-models).
-    diart
-    ```
 
-    Diart uses by default [pyannote.audio](https://github.com/pyannote/pyannote-audio) models from the _huggingface hub_. To use them, please follow the steps described [here](https://github.com/juanmc2005/diart?tab=readme-ov-file#get-access-to--pyannote-models).
+## Usage
+
+### Using the command-line tool
+
+After installation, you can start the server using the provided command-line tool:
+
+```bash
+whisperlivekit-server --host 0.0.0.0 --port 8000 --model tiny.en
+```
+
+Then open your browser at `http://localhost:8000` (or your specified host and port).
+
+### Using the library in your code
+
+```python
+from whisperlivekit import WhisperLiveKit
+from whisperlivekit.audio_processor import AudioProcessor
+from fastapi import FastAPI, WebSocket
+
+kit = WhisperLiveKit(model="medium", diarization=True)
+app = FastAPI() # Create a FastAPI application
+
+@app.get("/")
+async def get():
+    return HTMLResponse(kit.web_interface()) # Use the built-in web interface
+
+async def handle_websocket_results(websocket, results_generator): # Sends results to frontend
+    async for response in results_generator:
+        await websocket.send_json(response)
+
+@app.websocket("/asr")
+async def websocket_endpoint(websocket: WebSocket):
+    audio_processor = AudioProcessor()
+    await websocket.accept()
+    results_generator = await audio_processor.create_tasks()
+    websocket_task = asyncio.create_task(handle_websocket_results(websocket, results_generator))
+
+    while True:
+        message = await websocket.receive_bytes()
+        await audio_processor.process_audio(message)
+```
+
+For a complete audio processing example, check [whisper_fastapi_online_server.py](https://github.com/QuentinFuxa/WhisperLiveKit/blob/main/whisper_fastapi_online_server.py)
 
 
-3. **Run the FastAPI Server**:
+## Configuration Options
 
-    ```bash
+The following parameters are supported when initializing `WhisperLiveKit`:
-    python whisper_fastapi_online_server.py --host 0.0.0.0 --port 8000
-    ```
 
-    **Parameters**
+  - `--host` and `--port` let you specify the server's IP/port. 
-   
+  - `-min-chunk-size` sets the minimum chunk size for audio processing. Make sure this value aligns with the chunk size selected in the frontend. If not aligned, the system will work but may unnecessarily over-process audio data.
-    The following parameters are supported:
+  - `--transcription`: Enable/disable transcription (default: True)
-  
+  - `--diarization`: Enable/disable speaker diarization (default: False)
-    - `--host` and `--port` let you specify the server's IP/port. 
+  - `--confidence-validation`: Use confidence scores for faster validation. Transcription will be faster but punctuation might be less accurate (default: True)
-    - `-min-chunk-size` sets the minimum chunk size for audio processing. Make sure this value aligns with the chunk size selected in the frontend. If not aligned, the system will work but may unnecessarily over-process audio data.
+  - `--warmup-file`: The path to a speech audio wav file to warm up Whisper so that the very first chunk processing is fast. :
-    - `--transcription`: Enable/disable transcription (default: True)
+    - If not set, uses https://github.com/ggerganov/whisper.cpp/raw/master/samples/jfk.wav.
-    - `--diarization`: Enable/disable speaker diarization (default: False)
+    - If False, no warmup is performed.
-    - `--confidence-validation`: Use confidence scores for faster validation. Transcription will be faster but punctuation might be less accurate (default: True)
+  - `--min-chunk-size` Minimum audio chunk size in seconds. It waits up to this time to do processing. If the processing takes shorter time, it waits, otherwise it processes the whole segment that was received by this time.
-    - `--warmup-file`: The path to a speech audio wav file to warm up Whisper so that the very first chunk processing is fast. :
+  - `--model` {_tiny.en, tiny, base.en, base, small.en, small, medium.en, medium, large-v1, large-v2, large-v3, large, large-v3-turbo_}
-      - If not set, uses https://github.com/ggerganov/whisper.cpp/raw/master/samples/jfk.wav.
-      - If False, no warmup is performed.
-    - `--min-chunk-size` Minimum audio chunk size in seconds. It waits up to this time to do processing. If the processing takes shorter time, it waits, otherwise it processes the whole segment that was received by this time.
-    - `--model` {_tiny.en, tiny, base.en, base, small.en, small, medium.en, medium, large-v1, large-v2, large-v3, large, large-v3-turbo_}
                         Name size of the Whisper model to use (default: tiny). The model is automatically downloaded from the model hub if not present in model cache dir.
-    - `--model_cache_dir` Overriding the default model cache dir where models downloaded from the hub are saved
+  - `--model_cache_dir` Overriding the default model cache dir where models downloaded from the hub are saved
-    - `--model_dir` Dir where Whisper model.bin and other files are saved. This option overrides --model and --model_cache_dir parameter.
+  - `--model_dir` Dir where Whisper model.bin and other files are saved. This option overrides --model and --model_cache_dir parameter.
-    - `--lan`, --language Source language code, e.g. en,de,cs, or 'auto' for language detection.
+  - `--lan`, --language Source language code, e.g. en,de,cs, or 'auto' for language detection.
-    - `--task` {_transcribe, translate_} Transcribe or translate. If translate is set, we recommend avoiding the _large-v3-turbo_ backend, as it [performs significantly worse](https://github.com/QuentinFuxa/whisper_streaming_web/issues/40#issuecomment-2652816533) than other models for translation.
+  - `--task` {_transcribe, translate_} Transcribe or translate. If translate is set, we recommend avoiding the _large-v3-turbo_ backend, as it [performs significantly worse](https://github.com/QuentinFuxa/whisper_streaming_web/issues/40#issuecomment-2652816533) than other models for translation.
-    - `--backend` {_faster-whisper, whisper_timestamped, openai-api, mlx-whisper_} Load only this backend for Whisper processing.
+  - `--backend` {_faster-whisper, whisper_timestamped, openai-api, mlx-whisper_} Load only this backend for Whisper processing.
-    - `--vac` Use VAC = voice activity controller. Requires torch.
+  - `--vac` Use VAC = voice activity controller. Requires torch.
-    - `--vac-chunk-size` VAC sample size in seconds.
+  - `--vac-chunk-size` VAC sample size in seconds.
-    - `--vad` Use VAD = voice activity detection, with the default parameters.
+  - `--vad` Use VAD = voice activity detection, with the default parameters.
-    - `--buffer_trimming` {_sentence, segment_} Buffer trimming strategy -- trim completed sentences marked with punctuation mark and detected by sentence segmenter, or the completed segments returned by Whisper. Sentence segmenter must be installed for "sentence" option.
+  - `--buffer_trimming` {_sentence, segment_} Buffer trimming strategy -- trim completed sentences marked with punctuation mark and detected by sentence segmenter, or the completed segments returned by Whisper. Sentence segmenter must be installed for "sentence" option.
-    - `--buffer_trimming_sec` Buffer trimming length threshold in seconds. If buffer length is longer, trimming sentence/segment is triggered.
+  - `--buffer_trimming_sec` Buffer trimming length threshold in seconds. If buffer length is longer, trimming sentence/segment is triggered.
 
 5. **Open the Provided HTML**:
 
@@ -135,12 +150,13 @@ You need to install FFmpeg on your system:
     - Open your browser at `http://localhost:8000` (or replace `localhost` and `8000` with whatever you specified).  
     - The page uses vanilla JavaScript and the WebSocket API to capture your microphone and stream audio to the server in real time.
 
-### How the Live Interface Works
+
+## How the Live Interface Works
 
 - Once you **allow microphone access**, the page records small chunks of audio using the **MediaRecorder** API in **webm/opus** format.  
 - These chunks are sent over a **WebSocket** to the FastAPI endpoint at `/asr`.  
 - The Python server decodes `.webm` chunks on the fly using **FFmpeg** and streams them into the **whisper streaming** implementation for transcription.  
-- **Partial transcription** appears as soon as enough audio is processed. The “unvalidated” text is shown in **lighter or grey color** (i.e., an ‘aperçu’) to indicate it’s still buffered partial output. Once Whisper finalizes that segment, it’s displayed in normal text.  
+- **Partial transcription** appears as soon as enough audio is processed. The "unvalidated" text is shown in **lighter or grey color** (i.e., an 'aperçu') to indicate it's still buffered partial output. Once Whisper finalizes that segment, it's displayed in normal text.
 - You can watch the transcription update in near real time, ideal for demos, prototyping, or quick debugging.
 
 ### Deploying to a Remote Server

setup.py

@@ -28,7 +28,7 @@ setup(
     },
     entry_points={
         'console_scripts': [
-            'whisperlivekit-server=whisperlivekit.server:run_server',
+            'whisperlivekit-server=whisperlivekit.basic_server:main',
         ],
     },
     classifiers=[

whisperlivekit/basic_server.py

@@ -0,0 +1,86 @@
+from contextlib import asynccontextmanager
+from fastapi import FastAPI, WebSocket, WebSocketDisconnect
+from fastapi.responses import HTMLResponse
+from fastapi.middleware.cors import CORSMiddleware
+
+from whisperlivekit import WhisperLiveKit
+from whisperlivekit.audio_processor import AudioProcessor
+
+import asyncio
+import logging
+import os
+
+logging.basicConfig(level=logging.INFO, format="%(asctime)s - %(levelname)s - %(message)s")
+logging.getLogger().setLevel(logging.WARNING)
+logger = logging.getLogger(__name__)
+logger.setLevel(logging.DEBUG)
+
+kit = None
+
+@asynccontextmanager
+async def lifespan(app: FastAPI):
+    global kit
+    kit = WhisperLiveKit()
+    yield
+
+app = FastAPI(lifespan=lifespan)
+app.add_middleware(
+    CORSMiddleware,
+    allow_origins=["*"],
+    allow_credentials=True,
+    allow_methods=["*"],
+    allow_headers=["*"],
+)
+
+
+@app.get("/")
+async def get():
+    return HTMLResponse(kit.web_interface())
+
+
+async def handle_websocket_results(websocket, results_generator):
+    """Consumes results from the audio processor and sends them via WebSocket."""
+    try:
+        async for response in results_generator:
+            await websocket.send_json(response)
+    except Exception as e:
+        logger.warning(f"Error in WebSocket results handler: {e}")
+
+
+@app.websocket("/asr")
+async def websocket_endpoint(websocket: WebSocket):
+    audio_processor = AudioProcessor()
+
+    await websocket.accept()
+    logger.info("WebSocket connection opened.")
+            
+    results_generator = await audio_processor.create_tasks()
+    websocket_task = asyncio.create_task(handle_websocket_results(websocket, results_generator))
+
+    try:
+        while True:
+            message = await websocket.receive_bytes()
+            await audio_processor.process_audio(message)
+    except WebSocketDisconnect:
+        logger.warning("WebSocket disconnected.")
+    finally:
+        websocket_task.cancel()
+        await audio_processor.cleanup()
+        logger.info("WebSocket endpoint cleaned up.")
+
+def main():
+    """Entry point for the CLI command."""
+    import uvicorn
+    
+    temp_kit = WhisperLiveKit(transcription=False, diarization=False)
+    
+    uvicorn.run(
+        "whisperlivekit.basic_server:app", 
+        host=temp_kit.args.host, 
+        port=temp_kit.args.port, 
+        reload=True,
+        log_level="info"
+    )
+
+if __name__ == "__main__":
+    main()

whisperlivekit/core.py

@@ -1,4 +1,7 @@
-from whisperlivekit.whisper_streaming_custom.whisper_online import backend_factory, warmup_asr
+try:
+    from whisperlivekit.whisper_streaming_custom.whisper_online import backend_factory, warmup_asr
+except:
+    from whisper_streaming_custom.whisper_online import backend_factory, warmup_asr
 from argparse import Namespace, ArgumentParser
 
 def parse_args():