pyannote.audio

Use it when

• identify who spoke when in meetings, interviews, and podcasts
• add speaker diarization to an existing speech-to-text pipeline

What it solves

Not the fit when

• pretrained pipelines may require model terms acceptance and authentication

Install

pip install pyannote.audio

Invoke

Load a pretrained speaker-diarization pipeline, run it on the audio file, and merge speaker turns with transcript timestamps.

Alternatives

No reviewed alternatives recorded yet.