
Independent company · Paris, France
pyannoteAI
The creators of pyannote.audio, the open-source speaker diarization toolkit, now also offering hosted models through pyannoteAI.
Paris, France
- Founded
- 2024
- Type
- Independent company
- Based in
- Paris, France
- Key papers
- 3
- Featured projects
- 8
About
pyannote.audio was created by Hervé Bredin and collaborators, described in an ICASSP 2020 paper, and is released under the MIT licence. Its pretrained models are widely downloaded on Hugging Face.
pyannoteAI was founded in 2024 with Vincent Molina as CEO and Bredin as chief science officer. It sells faster hosted models such as precision-2.
In late 2025 the project released pyannote.audio 4.0 with the open community-1 model.
Mission and ambitionsThe team aims to give developers accurate speaker intelligence, keeping the core toolkit open while funding development through hosted models.
Known for
- 01
pyannote.audio
- 02
Speaker diarization
- 03
community-1 open model
- 04
Hugging Face pipelines
Milestones
- 2020
pyannote.audio paper
Published the toolkit at ICASSP.
- 2024
pyannoteAI founded
Launched the company around the toolkit.
- 2025
pyannote.audio 4.0
Released version 4.0 with community-1.
Open source
Neural building blocks for speaker diarization (10,806 stars)
Face detection, tracking and clustering in videos (477 stars)
Toolkit for reproducible diarization evaluation (258 stars)
Data structures for temporal segments with labels (126 stars)
Speaker diarization pipeline, the group's most downloaded model.
Community diarization pipeline.
Speaker segmentation model.
Speaker embedding model.
Research papers
Notable releases
Openness
Core code is MIT-licensed and several models are public on Hugging Face.
People
- Hervé BredinCreator of pyannote.audio and chief science officer
- Juan CoriaCTO (named by some sources)
Listed from the lab's own pages and public profiles.
Contact and community
The pyannoteAI contact page lists a support address and the community Discord.
Community server linked from the pyannoteAI site.
Sources
Compiled from the lab's public pages, Hugging Face and GitHub. Last checked 2026-11-17. Something wrong or out of date? Tell us on Discord.



