pyannoteAI, Gladia, and Modal are getting together for an evening on the engineering behind production voice AI.
Voice applications stack multiple layers: diarization, transcription, and the compute infrastructure underneath. We'll get into the trade-offs that show up when these layers meet real production workloads:
-
Designing the voice stack across speech intelligence and infrastructure
-
Managing latency, reliability, and cost
-
Getting from a working demo to something that holds up in production
This is for engineers and builders working on real voice products. Come with the problem you've been stuck on - there'll be plenty of people who've hit the same wall.
Agenda
-
7:00PM: Arrival: drinks, and mingle with other engineers
-
7:15PM: Panel & demo
-
7:50PM: Open Q&A
-
8:00PM: Food, drinks & more chat
