Model reference · open weights

overlapped-speech-detection

overlapped-speech-detection is an open-weight audio or speech model from pyannote.

• The overlapped-speech-detection model is developed by pyannote and is designed for the automatic speech recognition task.
• It identifies segments in audio where two or more speakers are active simultaneously.
• The model is released under the MIT license and requires pyannote.audio version 2.1 for installation and use.

Summary of the pyannote/overlapped-speech-detection model card, 2026-10-01

Hugging Face ↗