livekit-agents@1.6.9

svenstaro/genactlivekit-agents@1.6.9Aug 7, 2026by tinalenguyen

AI Summary

A feature release introducing Deepgram Flux TTS, Bland TTS, and various improvements to metrics, plugins (Phonic), and voice agent behavior. It also adds support for new models like OpenAI gpt-5.5.

Key Highlights

Added support for Deepgram's Flux TTS API (v2/speak)
Added Bland TTS plugin
Added Qwen3-ASR STT and Qwen3-TTS support via Baseten
Added OpenAI gpt-5.5 to model registry
Exposed expressive mode and added expressive_agent example

New Features

Deepgram Flux TTS API support
Bland TTS plugin
Baseten Qwen3 models
OpenAI gpt-5.5 support
Expressive mode implementation

Full Release Notes

## What's Changed
* Add support for Deepgram's Flux TTS API (v2/speak) integration by @dg-edcharbeneau in https://github.com/livekit/agents/pull/6511
* fix(cerebras): expose max_completion_tokens by @ryanl-cerebras in https://github.com/livekit/agents/pull/6685
* Make every example a uv workspace member by @u9g in https://github.com/livekit/agents/pull/6649
* feat(bland): add Bland TTS plugin by @rv-blandai in https://github.com/livekit/agents/pull/6554
* voice: forward agent_false_interruption over the RemoteSession by @theomonnom in https://github.com/livekit/agents/pull/6691
* plugins/phonic: expose more realtime config options by @arunwpm-work in https://github.com/livekit/agents/pull/6687
* fix(tts): give each fallback recovery path its own task slot by @LHMQ878 in https://github.com/livekit/agents/pull/6683
* tests: assert which session events the host forwards, not how many by @theomonnom in https://github.com/livekit/agents/pull/6693
* fix(livekit-agents): forward the STT hooks through MultiSpeakerAdapter by @LHMQ878 in https://github.com/livekit/agents/pull/6679
* fix(google): release the genai client when the LLM or Gemini TTS closes by @LHMQ878 in https://github.com/livekit/agents/pull/6686
* fix(exceptions): name the underlying cause when an LLM/TTS/STT connection fails by @u9g in https://github.com/livekit/agents/pull/6696
* feat(openai): add gpt-5.5 to model registry by @xtreme-sameer-vohra in https://github.com/livekit/agents/pull/6263
* feat(google): accept a credentials object on STT and TTS by @biztex in https://github.com/livekit/agents/pull/6618
* Feat/anam support ai disclosure by @sr-anam in https://github.com/livekit/agents/pull/6695
* fix(soniox): report RECOGNITION_USAGE on every frame by @vadimatmurphy in https://github.com/livekit/agents/pull/6585
* fix(voice): keep the run open while a tool holds the floor by @longcw in https://github.com/livekit/agents/pull/6694
* feat(metrics): broadcast llm tps + ttfs by @Bobronium in https://github.com/livekit/agents/pull/6373
* fix(amd): start SIP detection timeout after answer by @chenghao-mou in https://github.com/livekit/agents/pull/6580
* fix(deepgram): use punctuated_word for word-level transcripts by @mdylan2 in https://github.com/livekit/agents/pull/6699
* feat(eot): let a cloud TurnDetector opt out of the local fallback by @theomonnom in https://github.com/livekit/agents/pull/6719
* (expressive mode): remove extra space after stripping tag by @tinalenguyen in https://github.com/livekit/agents/pull/6717
* fix(audio): don't report a stopped playback as a decode error by @longcw in https://github.com/livekit/agents/pull/6723
* fix(aws): gate audio frames during text block sends to prevent Nova Sonic ValidationException by @lanazhang in https://github.com/livekit/agents/pull/6720
* feat(openai): support context hints in STT by @longcw in https://github.com/livekit/agents/pull/6705
* fix(deepgram): end_time should be the last word, not the first by @mdylan2 in https://github.com/livekit/agents/pull/6712
* fix(tts): probe a skipped provider for recovery on the streamed path by @biztex in https://github.com/livekit/agents/pull/6684
* fix(ipc): report actual consumed memory instead of RSS by @Bobronium in https://github.com/livekit/agents/pull/6082
* feat(voice): add session_host to AgentSession.start() by @theomonnom in https://github.com/livekit/agents/pull/6725
* fix(ipc): don't raise from is_alive after the inference process is closed by @RGB-loop in https://github.com/livekit/agents/pull/6732
* feat(baseten): add Qwen3-ASR STT and Qwen3-TTS support, and language_options for Whisper STT by @A-K-Erol in https://github.com/livekit/agents/pull/6700
* feat(fishaudio): enable quality-guard on all TTS requests by @cshape in https://github.com/livekit/agents/pull/6734
* (core): support duplicate tool policy specification by @tinalenguyen in https://github.com/livekit/agents/pull/6610
* fix(xai): emit one final user transcript per turn by @longcw in https://github.com/livekit/agents/pull/6729
* feat(inference): log gateway quota headers when LLM requests hit a 429 by @adrian-cowham in https://github.com/livekit/agents/pull/6736
* feat(expressive): expose expressive mode and add expressive_agent example by @theomonnom in https://github.com/livekit/agents/pull/6698
* livekit-agents@1.6.9 by @github-actions[bot] in https://github.com/livekit/agents/pull/6753

## New Contributors
* @dg-edcharbeneau made their first contribution in https://github.com/livekit/agents/pull/6511
* @rv-blandai made their first contribution in https://github.com/livekit/agents/pull/6554
* @LHMQ878 made their first contribution in https://github.com/livekit/agents/pull/6683
* @mdylan2 made their first contribution in https://github.com/livekit/agents/pull/6699
* @RGB-loop made their first contribution in https://github.com/livekit/agents/pull/6732
* @A-K-Erol made their first contribution in https://github.com/livekit/agents/pull/6700

**Full Changelog**: https://github.com/livekit/agents/compare/livekit-agents@1.6.8...livekit-agents@1.6.9