transcribe¶

class audinota.transcribe.TranscribeAudioResult(text: str, segments: list[faster_whisper.transcribe.Segment], info: faster_whisper.transcribe.TranscriptionInfo)[source]¶

audinota.transcribe.transcribe_audio_in_parallel(audio: BinaryIO, seg_duration: int = 120, n_jobs: int | None = None) → str[source]¶

Transcribe audio in parallel by splitting it into segments of specified duration.

Parameters:

audio – Audio data as a binary stream. Usually from io.BytesIO(Path("...").read_bytes()).
seg_duration – Duration of each segment in seconds. Default is 120 seconds.
n_jobs – Number of parallel jobs to run. If None, uses all available CPU cores.

Returns:

Transcribed text from the audio.