Hervé Bredin is Co-Founder and Chief Science Officer at pyannoteAI, pioneering speaker intelligence technology that identifies who speaks when in audio recordings. He dedicated over a decade to developing the world's most widely adopted speaker diarization toolkit, transforming how AI processes voice data.

Hervé received his PhD on talking-face biometric authentication from Telecom ParisTech in 2007. He has been a permanent researcher at CNRS since 2008, working almost exclusively on speaker diarization and related tasks, and was a member of the Institut de Recherche en Informatique de Toulouse (IRIT).

Known as the godfather of diarization, Hervé spent a decade refining speaker segmentation and his open-source toolkit, pyannote-audio, became the global gold standard. The toolkit has reached 8k+ GitHub followers, 140k unique registered users, and 45 million monthly downloads on Hugging Face.

pyannoteAI raised $9M in Seed funding led by Crane Venture Partners and Serena, with backing from HuggingFace CTO, former Meta and OpenAI researchers. The company was founded in 2024 by Hervé Bredin, Vincent Molina, and Juan Coria.

The open-source foundation powers 100,000+ developers worldwide with 45 million monthly downloads on HuggingFace. As the foundational layer of conversational AI, speaker diarization unlocks applications including meeting transcription, call center analytics, voice agents, and video dubbing.