quail.decode_speech¶
- quail.decode_speech(path, model_size='base', save=False, return_raw=False, **kwargs)[source]¶
Decode speech for a file or folder and return results using OpenAI Whisper.
- Parameters:
- pathstr
Path to a wav file, or a folder of wav files.
- model_sizestr
Whisper model size: ‘tiny’, ‘base’, ‘small’, ‘medium’, ‘large’. Default is ‘base’.
- saveboolean
False by default. If true, saves results object (pickle) and text transcript.
- return_rawboolean
If True, returns the full Whisper result dictionary. If False (default), returns a list of (WORD, START, END) tuples.
- **kwargsdict
Additional arguments passed to whisper.transcribe (e.g. language).
- Returns:
- wordslist of str, or list of lists of str
The results of the speech decoding.