What's Changed
- fix(llm): tolerate unparseable stored tool-call arguments in Anthropic/Google/AWS formatters by @PranavMishra28 in #6321
- aws: omit temperature/topP for Bedrock models that reject them by @rrfunde in #6947
- fix(tests): settle genai client finalizers in the test that owns them by @helin-mktech in #6882
- fix(gradium): update parameters by @timpratim in #7014
- fix(xai): reject FileSearch plus collections_search or file_search by @mehrdroid in #6998
- fix(aws): import AsyncBedrockRuntimeClient for bedrock-runtime 0.10+ by @libaojiang in #7002
- chore: remove unused segment_uuid and speech_uuid helpers by @detail-app[bot] in #7021
- fix: handle RoomIO microphone track replacement by @chenghao-mou in #6744
- fix(openai): handle incomplete Responses events by @dawNotPoi in #7024
- fix: correctly handle SIGTERM/SIGINT even when sync functions block by @davidzhao in #7034
- avoid shutting down third-party otel providers by @davidzhao in #7033
- fix(mistralai): support assistant-ending chat contexts by @mikemikimike in #7036
- feat(deepgram): add numerals, profanity_filter, and redact options to STTv2 (Flux) by @vadimatmurphy in #6993
- fix(openai): close the STT's websocket connection pool by @Rehansanjay in #7023
- fix: retry connection pool prewarm after failure by @jstar0 in #6969
- fix(plugins): redact API keys from rejected websocket handshakes by @aniketwaghh in #7032
- fix(elevenlabs): don't forward re-sent partial transcripts by @captainbanan in #6922
- fix: transcription_delay reported as ~0 with the default VAD by @davidzhao in #6557
- feat(ultravox): implement say() via ForcedAgentMessage by @ssgutierrez42 in #7001
- fix: disable default AGC with noise cancellation by @chenghao-mou in #7064
- refactor(hotel_receptionist): split scenarios into scoped and open-ended files by @u9g in #7065
- feat(google): support custom prompt config for STT chirp_3 by @kechako in #7055
- fix(simplismart): close the streams the STT collects by @Rehansanjay in #7012
- feat(metrics): expose reasoning_tokens in LLM metrics and traces by @s-hamdananwar in #7068
- docs: add --audio_eot to documented CI unit gate command by @detail-app[bot] in #7070
- fix(voice): flush recorder resampler tails at run end by @dajiaohuang in #7047
- fix(aws): pin the CRT transport for Nova Sonic by @longcw in #7076
- fix(voice): cancel parked preemptive generation before pausing the activity by @uuzzrm in #6865
- fix(avatar): recover DataStreamAudioReceiver from a lost stream trailer by @jasper-face-up in #7007
- Bump livekit 1.1.17 by @cnderrauber in #7078
- feat: twilio connector warm transfer by @anunaym14 in #6283
- fix: resolve transcript-only amd greetings at eot by @chenghao-mou in #7026
- fix(cartesia): ignore stale frames from previous contexts on pooled websockets by @Panmax in #7081
- feat(elevenlabs): add secondary_languages for Scribe v2 realtime STT by @tinalenguyen in #7086
- fix(llm): keep falsy tool return values in make_function_call_output by @Osamaali313 in #6551
- fix(google): don't let closing the TTS input generator mask the real error by @avionicharshit-byte in #7041
- fix(aws): restore Transcribe STT on transcribe-streaming 0.11 by @longcw in #7094
- fix(google): route Gemma 4 thinking through thinking_level by @longcw in #7095
- fix(inworld): keep a reference to the prewarm task and cancel it on close by @Rehansanjay in #7050
- chore(deps): update dependency nltk to v3.10.3 [security] by @renovate[bot] in #7091
- feat(assemblyai): add universal-3-6-pro streaming model by @dlange-aai in #7100
- Remove deprecated Hedra avatar plugin by @bcherry in #7084
- feat(phonic): add update_options for mid-session config changes by @arunwpm-work in #7089
- chore(lint): enable RUF006 so dangling asyncio tasks fail CI by @Rehansanjay in #7054
- tests: stop room tests hanging on macOS by @longcw in #6857
- feat: add Vakyam TTS plugin by @Adithya-Sakaray in #7009
- fix(hotel_receptionist): stop a restaurant modification silently resizing the party by @u9g in #6806
- chore(inference): remove ElevenLabs models after the gateway brownout by @chenghao-mou in #7106
- feat(assemblyai): surface end_of_turn_confidence on SpeechData.metadata by @gsharp-aai in #7107
- fix(voice): stop the recorder writing over audio still in the resampler by @longcw in #7079
- fix: timestamp say transcripts at speech start by @chenghao-mou in #7110
- fix: restore endpointing overlap when agent speech resumes by @chenghao-mou in #6939
- fix: exclude agent speech from dynamic endpointing pauses by @chenghao-mou in #7061
- fix: preserve forkserver copy-on-write sharing by @chenghao-mou in #7120
- feat(phonic): export LiveKit tools for the Responses API by @elizabeth-phonic in #7112
- fix: release stale transcript gates and playout pauses by @chenghao-mou in #7083
- fix: restore speaking state when releasing playout pause by @chenghao-mou in #7087
- feat(telemetry): full OTel GenAI semantic conventions + in-process PII stripping by @theomonnom in #7104
- livekit-agents@1.8.0 by @github-actions[bot] in #7123
New Contributors
- @PranavMishra28 made their first contribution in #6321
- @rrfunde made their first contribution in #6947
- @helin-mktech made their first contribution in #6882
- @timpratim made their first contribution in #7014
- @mehrdroid made their first contribution in #6998
- @libaojiang made their first contribution in #7002
- @dawNotPoi made their first contribution in #7024
- @Rehansanjay made their first contribution in #7023
- @jstar0 made their first contribution in #6969
- @aniketwaghh made their first contribution in #7032
- @dajiaohuang made their first contribution in #7047
- @jasper-face-up made their first contribution in #7007
- @Osamaali313 made their first contribution in #6551
- @avionicharshit-byte made their first contribution in #7041
- @Adithya-Sakaray made their first contribution in #7009
- @elizabeth-phonic made their first contribution in #7112
Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.7.1...livekit-agents@1.8.0