About this event
Live-1, our streaming diarization model is here.
Join us for a live, technical walkthrough of the most accurate streaming diarization model on the market.
Real-time speaker intelligence, without compromising on accuracy.
Voice AI has a latency problem. Most diarization systems were built for batch: process, wait, get results. But real-world voice applications can't afford to wait. They need to know who's speaking now.
Join Hervé Bredin, Chief Science Officer, Juan Coria, Chief Technology Officer, and Thomas Mol, Founding Engineer, for a live introduction and hands-on tutorial covering everything you need to get started with Streaming Diarization.
We'll walk you through:
- Why Voice AI pipelines need real-time speaker intelligence
- The mission behind Streaming Diarization: Bring batch-level accuracy at the speed of real-time
- A full product walkthrough: Architecture, configuration, and output format
- A live playground demonstration: From setup to JSON interpretation
- API integration: How to plug Streaming Diarization into a Voice AI pipeline
- A step-by-step tutorial: Building a live subtitling system using pyannoteAI + an STT provider
- Q&A with the team
When? 📅 July 7, 2026 | 5 PM CEST
Don’t miss this chance to see how Streaming Diarization works end-to-end, and leave with everything you need to start building a live Voice AI stack.
pyannote.ai
Identify who speaks when with pyannote.aiSimply detect, segment, label, and separate speakers in any language.
When your audio sources are complex, contaminated by background noise, affected by external elements, and the performances of your solution depend on them, pyannote brings accuracy and precision.
