Model reference · open weights

speaker-segmentation

speaker-segmentation is an open-weight audio or speech model from pyannote.

• The speaker-segmentation model is an open-source pipeline developed by pyannote for automatic speech recognition tasks.
• It identifies speaker turns in audio files using the pyannote.audio 2.1 framework and is distributed under the MIT license.
• The card notes that this specific pipeline does not address speaker diarization.

Summary of the pyannote/speaker-segmentation model card, 2026-10-01

Hugging Face ↗