Starstruk Logo MarkStarstruk

livekit-agents@1.6.10

svenstaro/genactlivekit-agents@1.6.10Aug 20, 2026by tinalenguyen

AI Summary

A patch release addressing voice agent stability issues related to interruptions, recording uploads, and session management. It adds fixes for Azure TTS and exposes new metrics for cache creation tokens.

Key Highlights

  • Fixed interrupt walk behavior to stop at queued speech disallowing interruptions
  • Fixed recording upload connection failures
  • Added Azure TTS model string replacement (unknown -> host/region + voice)
  • Exposed cache_creation_tokens in LLM metrics
  • Added recording options to session reports

New Features

  • Azure TTS model string fix
  • Cache creation tokens metrics
  • Session recording options
  • Low inference priority for judges and simulations

Full Release Notes

## What's Changed
* fix(cartesia): redact API keys from websocket handshake errors by @LHMQ878 in https://github.com/livekit/agents/pull/6740
* fix: respect non-retryable STT errors by @chenghao-mou in https://github.com/livekit/agents/pull/6759
* fix: fallback adapters report the active instance's model/provider by @Fyko in https://github.com/livekit/agents/pull/6690
* fix(inference): don't crash the session on null usage token counts by @biztex in https://github.com/livekit/agents/pull/6763
* fix(voice): stop the interrupt walk at a queued speech that disallows interruptions by @biztex in https://github.com/livekit/agents/pull/6644
* fix(voice): don't raise when interrupting a cancelled or finished speech by @longcw in https://github.com/livekit/agents/pull/6767
* fix: isolate bundled VAD inference from the default executor by @chenghao-mou in https://github.com/livekit/agents/pull/6770
* fix(xai): align realtime Voice Agent defaults and API coverage by @damien-xai in https://github.com/livekit/agents/pull/6755
* feat(spatius): default audio format to ogg_opus by @3DRX in https://github.com/livekit/agents/pull/6749
* fix: don't count discarded room audio as played by @chenghao-mou in https://github.com/livekit/agents/pull/6728
* feat(azure): replace unknown tts model string with host/region + voice by @alvxrog in https://github.com/livekit/agents/pull/6775
* feat: expose cache_creation_tokens in LLM metrics by @chasleslr in https://github.com/livekit/agents/pull/6655
* fix: treat a silent agent turn as silence, not a run_input error by @theomonnom in https://github.com/livekit/agents/pull/6779
* fix: stop dropping session host responses by @theomonnom in https://github.com/livekit/agents/pull/6781
* fix(llm): never give a const field the null default sentinel by @u9g in https://github.com/livekit/agents/pull/6688
* fix: hold pending replies at playout launch by @chenghao-mou in https://github.com/livekit/agents/pull/6745
* fix(inference): keep a stream retryable until it emits generation by @u9g in https://github.com/livekit/agents/pull/6697
* test: stabilize text email collection eval by @chenghao-mou in https://github.com/livekit/agents/pull/6791
* feat(evals): run judges at low inference priority by @u9g in https://github.com/livekit/agents/pull/6794
* feat(simulation): pin text simulations to the low inference class by @theomonnom in https://github.com/livekit/agents/pull/6721
* fix(google): allow explicit project for Speech-to-Text by @uuzzrm in https://github.com/livekit/agents/pull/6810
* fix(silero): update bundled VAD checkpoint to v6.2 by @longcw in https://github.com/livekit/agents/pull/6816
* fix(voice): export uncommitted user_turn spans on close by @chenghao-mou in https://github.com/livekit/agents/pull/6592
* feat(telemetry): include recording options in session reports by @chenghao-mou in https://github.com/livekit/agents/pull/6547
* fix: respect agent endpointing delay during reply hold-off by @chenghao-mou in https://github.com/livekit/agents/pull/6820
* docs: require confirmation for non-patch version bumps by @chenghao-mou in https://github.com/livekit/agents/pull/6821
* examples: switch homepage agent to Fish Audio with expressive mode by @theomonnom in https://github.com/livekit/agents/pull/6822
* fix: stabilize LiveKit pronunciation with Fish Audio by @chenghao-mou in https://github.com/livekit/agents/pull/6825
* fix(deps): pin openai below 3 by @longcw in https://github.com/livekit/agents/pull/6831
* livekit-agents@1.6.10 by @github-actions[bot] in https://github.com/livekit/agents/pull/6834

## New Contributors
* @Fyko made their first contribution in https://github.com/livekit/agents/pull/6690
* @damien-xai made their first contribution in https://github.com/livekit/agents/pull/6755
* @alvxrog made their first contribution in https://github.com/livekit/agents/pull/6775
* @chasleslr made their first contribution in https://github.com/livekit/agents/pull/6655

**Full Changelog**: https://github.com/livekit/agents/compare/livekit-agents@1.6.9...livekit-agents@1.6.10