diff --git a/README.md b/README.md index 82c03ef..9f3a394 100644 --- a/README.md +++ b/README.md @@ -2,7 +2,7 @@
Updates: -21 Mar 2024: Added a 'wizard' into the launcher that will help standalone users get common Bazarr variables configured. See below in Launcher section. +21 Mar 2024: Added a 'wizard' into the launcher that will help standalone users get common Bazarr variables configured. See below in Launcher section. Removed 'Transformers' as an option. While I usually don't like to remove features, I don't think anyone is using this and the results are wildly unpredictable and often cause out of memory errors. 19 Mar 2024: Added a `MONITOR` environment variable. Will 'watch' or 'monitor' your `TRANSCRIBE_FOLDERS` for changes and run on them. Useful if you just want to paste files into a folder and get subtitles. @@ -169,8 +169,6 @@ The following environment variables are available in Docker. They will default | DEBUG | True | Provides some debug data that can be helpful to troubleshoot path mapping and other issues. Fun fact, if this is set to true, any modifications to the script will auto-reload it (if it isn't actively transcoding). Useful to make small tweaks without re-downloading the whole file. | | FORCE_DETECTED_LANGUAGE_TO | '' | This is to force the model to a language instead of the detected one, takes a 2 letter language code. For example, your audio is French but keeps detecting as English, you would set it to 'fr' | | CLEAR_VRAM_ON_COMPLETE | True | This will delete the model and do garbage collection when queue is empty. Good if you need to use the VRAM for something else. | -| HF_TRANSFORMERS | False | Uses Hugging Face Transformers models that should be faster, not tested as of now because HF is down. | -| HF_BATCH_SIZE | 24 | Batch size to be used with above. Batch size has a correlation to VRAM, not sure what it is yet and may require tinkering. | UPDATE | False | Will pull latest subgen.py from the repository if True. False will use the original subgen.py built into the Docker image. Standalone users can use this with launcher.py to get updates. | | APPEND | False | Will add the following at the end of a subtitle: "Transcribed by whisperAI with faster-whisper ({whisper_model}) on {datetime.now()}" | MONITOR | False | Will monitor `TRANSCRIBE_FOLDERS` for real-time changes to see if we need to generate subtitles | diff --git a/requirements.txt b/requirements.txt index 44d9d7a..2392686 100644 --- a/requirements.txt +++ b/requirements.txt @@ -7,7 +7,4 @@ uvicorn python-multipart python-ffmpeg whisper -transformers -accelerate -optimum watchdog diff --git a/subgen.py b/subgen.py index cebe19f..355796c 100644 --- a/subgen.py +++ b/subgen.py @@ -53,8 +53,6 @@ monitor = convert_to_bool(os.getenv('MONITOR', False)) transcribe_folders = os.getenv('TRANSCRIBE_FOLDERS', '') transcribe_or_translate = os.getenv('TRANSCRIBE_OR_TRANSLATE', 'transcribe') force_detected_language_to = os.getenv('FORCE_DETECTED_LANGUAGE_TO', '') -hf_transformers = convert_to_bool(os.getenv('HF_TRANSFORMERS', False)) -hf_batch_size = int(os.getenv('HF_BATCH_SIZE', 24)) clear_vram_on_complete = convert_to_bool(os.getenv('CLEAR_VRAM_ON_COMPLETE', True)) compute_type = os.getenv('COMPUTE_TYPE', 'auto') append = convert_to_bool(os.getenv('APPEND', False)) @@ -334,10 +332,7 @@ def asr( start_model() files_to_transcribe.insert(0, f"Bazarr-asr-{random_name}") audio_data = np.frombuffer(audio_file.file.read(), np.int16).flatten().astype(np.float32) / 32768.0 - if(hf_transformers): - result = model.transcribe(audio_data, task=task, input_sr=16000, language=language, batch_size=hf_batch_size, progress_callback=progress) - else: - result = model.transcribe_stable(audio_data, task=task, input_sr=16000, language=language, progress_callback=progress) + result = model.transcribe_stable(audio_data, task=task, input_sr=16000, language=language, progress_callback=progress) appendLine(result) elapsed_time = time.time() - start_time minutes, seconds = divmod(int(elapsed_time), 60) @@ -370,10 +365,7 @@ def detect_language( random_name = random.choices("abcdefghijklmnopqrstuvwxyzABCDEFGHIJKLMNOPQRSTUVWXYZ1234567890", k=6) files_to_transcribe.insert(0, f"Bazarr-detect-language-{random_name}") audio_data = np.frombuffer(audio_file.file.read(), np.int16).flatten().astype(np.float32) / 32768.0 - if(hf_transformers): - detected_lang_code = model.transcribe(whisper.pad_or_trim(audio_data), input_sr=16000, batch_size=hf_batch_size).language - else: - detected_lang_code = model.transcribe_stable(whisper.pad_or_trim(audio_data), input_sr=16000).language + detected_lang_code = model.transcribe_stable(whisper.pad_or_trim(audio_data), input_sr=16000).language except Exception as e: logging.info(f"Error processing or transcribing Bazarr {audio_file.filename}: {e}") @@ -389,11 +381,7 @@ def start_model(): global model if model is None: logging.debug("Model was purged, need to re-create") - if(hf_transformers): - logging.debug("Using Hugging Face Transformers, whisper_threads, concurrent_transcriptions, and model_location variables are ignored!") - model = stable_whisper.load_hf_whisper(whisper_model, device=transcribe_device) - else: - model = stable_whisper.load_faster_whisper(whisper_model, download_root=model_location, device=transcribe_device, cpu_threads=whisper_threads, num_workers=concurrent_transcriptions, compute_type=compute_type) + model = stable_whisper.load_faster_whisper(whisper_model, download_root=model_location, device=transcribe_device, cpu_threads=whisper_threads, num_workers=concurrent_transcriptions, compute_type=compute_type) def delete_model(): if clear_vram_on_complete and len(files_to_transcribe) == 0: @@ -444,10 +432,7 @@ def gen_subtitles(file_path: str, transcribe_or_translate: str, front=True, forc if force_detected_language_to: forceLanguage = force_detected_language_to logging.info(f"Forcing language to {forceLanguage}") - if(hf_transformers): - result = model.transcribe(file_path, language=forceLanguage, batch_size=hf_batch_size, task=transcribe_or_translate, progress_callback=progress) - else: - result = model.transcribe_stable(file_path, language=forceLanguage, task=transcribe_or_translate, progress_callback=progress) + result = model.transcribe_stable(file_path, language=forceLanguage, task=transcribe_or_translate, progress_callback=progress) appendLine(result) result.to_srt_vtt(get_file_name_without_extension(file_path) + subextension, word_level=word_level_highlight) elapsed_time = time.time() - start_time @@ -772,10 +757,7 @@ if __name__ == "__main__": logging.info(f"Transcriptions are limited to running {str(concurrent_transcriptions)} at a time") logging.info(f"Running {str(whisper_threads)} threads per transcription") logging.info(f"Using {transcribe_device} to encode") - if hf_transformers: - logging.info(f"Using Hugging Face Transformers") - else: - logging.info(f"Using faster-whisper") + logging.info(f"Using faster-whisper") if transcribe_folders: transcribe_existing(transcribe_folders) uvicorn.run("subgen:app", host="0.0.0.0", port=int(webhookport), reload=reload_script_on_change, use_colors=True)