Model reference · open weights

speaker-diarization-community-1

speaker-diarization-community-1 is an open-weight audio or speech model from pyannote. speaker-diarization-community-1 (BF16) weighs 33 MB; the smallest configuration that runs it is RTX 3060 12 GB.

• speaker-diarization-community-1 is a pipeline by pyannote that processes mono audio sampled at 16kHz to output speaker diarization.
• It automatically downmixes multi-channel audio and resamples files to the required rate.
• The model is licensed under cc-by-4.0 and supports offline use.

Summary of the pyannote/speaker-diarization-community-1 model card, 2026-10-01

Hugging Face ↗Paper ↗