News for hackers. Stuff that matters.

A comment on How OpenAI delivers low-latency voice AI at scale.

Re:How OpenAI delivers low-latency voice AI at scale by charisma123 on Monday May 04, 2026 @04:48PM (#48014765)
If a transceiver crashes during a stream, how is the active session recovered? Does the system automatically re-establish the context in a new WebRTC session?
[ Parent | on HN ]
Re:How OpenAI delivers low-latency voice AI at scale by Sean-Der on Monday May 04, 2026 @04:53PM (#48014825)
It doesn't today, but you could with sometime like this [0]. You can save/suspend all WebRTC state and bring it back with the next process.

[0] https://github.com/pion/webrtc-zero-downtime-restart

[ Parent | on HN ]