Fact · Transcription
The pyannote.audio paper describes an open-source toolkit written in Python for speaker diarization, with pre-trained models for voice activity detection, speaker change detection, overlapped speech detection and speaker embedding.
Checked against its source on October 17, 2026
In the source's own words
“an open-source toolkit written in Python for speaker diarization”
pyannote.audio: neural building blocks for speaker diarization, Hervé Bredin et al., arXiv (submitted to ICASSP 2020), 2019, Abstract.
Open the original- What it measures
- The toolkit is built on the PyTorch machine learning framework. Diarization is split into separate trainable parts, and the paper says the pre-trained models cover voice activity detection, speaker change detection, overlapped speech detection and speaker embedding.
- What it does not show
- The abstract describes the design and the pre-trained parts of the toolkit. It does not give a single accuracy figure for a complete pipeline on any one recording.