Transcription

whisper_smith.transcribe.from_openai_response(response)[source]
Parameters:

response (OpenAITranscriptionResponse)

Return type:

TranscriptResult

whisper_smith.transcribe.transcribe_audio(audio_path, *, client=None, model='gpt-4o-transcribe', require_timestamps=True, timestamp_model='whisper-1')[source]
Parameters:

  • audio_path (str | Path)

  • client (OpenAI | None)

  • model (str)

  • require_timestamps (bool)

  • timestamp_model (str)

Return type:

TranscriptResult

whisper_smith.transcribe.transcribe_file(audio_path, *, client=None, model='gpt-4o-transcribe', require_timestamps=True, timestamp_model='whisper-1')[source]
Parameters:

  • audio_path (str | Path)

  • client (OpenAI | None)

  • model (str)

  • require_timestamps (bool)

  • timestamp_model (str)

Return type:

TranscriptResult