diff --git a/README.md b/README.md index 26d4d6a..82685ff 100644 --- a/README.md +++ b/README.md @@ -136,26 +136,26 @@ For a complete audio processing example, check [whisper_fastapi_online_server.py The following parameters are supported when initializing `WhisperLiveKit`: - `--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. - - `--transcription`: Enable/disable transcription (default: True) - - `--diarization`: Enable/disable speaker diarization (default: False) - - `--confidence-validation`: Use confidence scores for faster validation. Transcription will be faster but punctuation might be less accurate (default: True) - - `--warmup-file`: The path to a speech audio wav file to warm up Whisper so that the very first chunk processing is fast. : + - `--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. + - `--no-transcription`: Disable transcription (enabled by default) + - `--diarization`: Enable speaker diarization (disabled by default) + - `--confidence-validation`: Use confidence scores for faster validation. Transcription will be faster but punctuation might be less accurate (disabled by default) + - `--warmup-file`: The path to a speech audio wav file to warm up Whisper so that the very first chunk processing is fast: - 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_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. - - `--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. - - `--vac` Use VAC = voice activity controller. Requires torch. - - `--vac-chunk-size` VAC sample size in seconds. - - `--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_sec` Buffer trimming length threshold in seconds. If buffer length is longer, trimming sentence/segment is triggered. + - `--model`: Name size of the Whisper model to use (default: tiny). Suggested values: tiny.en, tiny, base.en, base, small.en, small, medium.en, medium, large-v1, large-v2, large-v3, large, large-v3-turbo. 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_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. + - `--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. + - `--vac`: Use VAC = voice activity controller. Requires torch. (disabled by default) + - `--vac-chunk-size`: VAC sample size in seconds. + - `--no-vad`: Disable VAD (voice activity detection), which is enabled by default. + - `--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. + ## How the Live Interface Works diff --git a/demo.png b/demo.png index dd3a9c4..ae46bf9 100644 Binary files a/demo.png and b/demo.png differ diff --git a/whisperlivekit/core.py b/whisperlivekit/core.py index 5d1e266..c0185d9 100644 --- a/whisperlivekit/core.py +++ b/whisperlivekit/core.py @@ -29,23 +29,21 @@ def parse_args(): parser.add_argument( "--confidence-validation", - type=bool, - default=False, + action="store_true", help="Accelerates validation of tokens using confidence scores. Transcription will be faster but punctuation might be less accurate.", ) parser.add_argument( "--diarization", - type=bool, - default=True, - help="Whether to enable speaker diarization.", + action="store_true", + default=False, + help="Enable speaker diarization.", ) parser.add_argument( - "--transcription", - type=bool, - default=True, - help="To disable to only see live diarization results.", + "--no-transcription", + action="store_true", + help="Disable transcription to only see live diarization results.", ) parser.add_argument( @@ -54,15 +52,14 @@ def parse_args(): default=0.5, help="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.", ) + parser.add_argument( "--model", type=str, default="tiny", - choices="tiny.en,tiny,base.en,base,small.en,small,medium.en,medium,large-v1,large-v2,large-v3,large,large-v3-turbo".split( - "," - ), - help="Name size of the Whisper model to use (default: large-v2). The model is automatically downloaded from the model hub if not present in model cache dir.", + help="Name size of the Whisper model to use (default: tiny). Suggested values: tiny.en,tiny,base.en,base,small.en,small,medium.en,medium,large-v1,large-v2,large-v3,large,large-v3-turbo. The model is automatically downloaded from the model hub if not present in model cache dir.", ) + parser.add_argument( "--model_cache_dir", type=str, @@ -105,12 +102,13 @@ def parse_args(): parser.add_argument( "--vac-chunk-size", type=float, default=0.04, help="VAC sample size in seconds." ) + parser.add_argument( - "--vad", + "--no-vad", action="store_true", - default=True, - help="Use VAD = voice activity detection, with the default parameters.", + help="Disable VAD (voice activity detection).", ) + parser.add_argument( "--buffer_trimming", type=str, @@ -134,6 +132,12 @@ def parse_args(): ) args = parser.parse_args() + + args.transcription = not args.no_transcription + args.vad = not args.no_vad + delattr(args, 'no_transcription') + delattr(args, 'no_vad') + return args class WhisperLiveKit: