livekit-agents@1.8.2
AudarAI/Audar-ASR-V1livekit-agents@1.8.2Sep 15, 2026by tinalenguyen
Upgrading?
AI Summary
Significant performance and telemetry improvements for agent lifecycle management, including blocking detection and faster startup times.
Key Highlights
- Added run loop blocking detection to surface stalls in Agent Insights
- Faster agent startup by up to 800ms via prefork server preloading
- Improved agent lifecycle tracing covering dispatch latency to session live
- Fixed Deepgram STT silently dropping sockets instead of hanging
- Added ElevenLabs secondary languages and language detection
New Features
- Run loop blocking detection
- Faster agent startup
- Enhanced telemetry spans
- ElevenLabs secondary languages
- Synthesia interactive-avatar plugin
Full Release Notes
### Optimizations - **Run loop blocking detection.** The agent now detects code that blocks the asyncio event loop and surfaces the offending call in Agent Insights, so you can find stalls without attaching a profiler. - **Faster agent startup.** The prefork server preloads libraries before forking, cutting agent startup time by up to 800ms. - **Improved agent lifecycle tracing.** Traces now cover the full agent lifecycle, including initial dispatch latency, making it clear where time goes between a job request and a session going live. ## Full changelog * feat(speechify): send Speechify-Caller-Version attribution header by @luke-speechify in https://github.com/livekit/agents/pull/7166 * fix(llm): reposition mid-conversation instructions for Gemini gateway and Mistral by @L65FREAD in https://github.com/livekit/agents/pull/6591 * feat(voice): log which participant RoomIO links to by @itsnicjohn in https://github.com/livekit/agents/pull/7210 * chore(openai): remove dead VertexModels type alias by @detail-app[bot] in https://github.com/livekit/agents/pull/7223 * fix: refuse to start a DuplexModel under a text simulation by @u9g in https://github.com/livekit/agents/pull/7236 * (deepgram stt): detect a silently dropped socket instead of hanging by @GregHolmes in https://github.com/livekit/agents/pull/7206 * fix(deps): pin anthropic below 1 by @omChauhanDev in https://github.com/livekit/agents/pull/7242 * Add Synthesia interactive-avatar plugin by @swarnimkulkarni-synth in https://github.com/livekit/agents/pull/7216 * feat(telemetry): trace the end-of-turn wait and the turn handoff by @davidzhao in https://github.com/livekit/agents/pull/7130 * feat(telemetry): trace dispatch, startup, and shutdown by @davidzhao in https://github.com/livekit/agents/pull/7131 * feat(telemetry): trace RPC calls made and handled by @davidzhao in https://github.com/livekit/agents/pull/7134 * feat(telemetry): interruption detail, handoff span, fallback events, text input by @davidzhao in https://github.com/livekit/agents/pull/7137 * feat(telemetry): one agent_turn span per speech handle by @davidzhao in https://github.com/livekit/agents/pull/7143 * tests: trace-shape schema and checker by @davidzhao in https://github.com/livekit/agents/pull/7148 * feat(elevenlabs): secondary languages and language detection for realtime STT by @captainbanan in https://github.com/livekit/agents/pull/6902 * fix(aio): drain blocked senders when closing channels by @Kuang-xianxin in https://github.com/livekit/agents/pull/7257 * fix(elevenlabs): remove the duplicated secondary_languages parameter by @longcw in https://github.com/livekit/agents/pull/7262 * fix(duplex): keep late transcripts on the audio they describe by @longcw in https://github.com/livekit/agents/pull/7264 * fix(room_io): frame output audio by a cap, not a fixed size by @longcw in https://github.com/livekit/agents/pull/7261 * fix(cli): prebuffer console playback so realtime-paced audio doesn't underrun by @u9g in https://github.com/livekit/agents/pull/7238 * fix(openai): ignore incomplete and duplicate GPT Live backend tool calls by @kiwijuicer in https://github.com/livekit/agents/pull/7230 * ipc: add room_id to job logging_extra by @andrewnitu in https://github.com/livekit/agents/pull/7241 * feat(rime): support Coda streaming in LiveKit by @naszzz in https://github.com/livekit/agents/pull/6978 * Migrate Speechmatics STT plugin to Agent STT by @Smiljanic19A in https://github.com/livekit/agents/pull/7028 * fix(openai): continue GPT Live only once every backend call is answered by @longcw in https://github.com/livekit/agents/pull/7266 * Revert "fix(room_io): frame output audio by a cap, not a fixed size (#7261)" by @longcw in https://github.com/livekit/agents/pull/7271 * fix(room_io): release a partial frame before the source runs dry by @longcw in https://github.com/livekit/agents/pull/7272 * fix: prevent llm fallback retries after output by @chenghao-mou in https://github.com/livekit/agents/pull/7233 * fix: reject inline tasks during agent handoff by @chenghao-mou in https://github.com/livekit/agents/pull/7268 * (gemini live): add new models by @tinalenguyen in https://github.com/livekit/agents/pull/7289 * livekit-agents@1.8.2 by @github-actions[bot] in https://github.com/livekit/agents/pull/7292 ## New Contributors * @L65FREAD made their first contribution in https://github.com/livekit/agents/pull/6591 * @GregHolmes made their first contribution in https://github.com/livekit/agents/pull/7206 * @swarnimkulkarni-synth made their first contribution in https://github.com/livekit/agents/pull/7216 * @kiwijuicer made their first contribution in https://github.com/livekit/agents/pull/7230 * @andrewnitu made their first contribution in https://github.com/livekit/agents/pull/7241 * @Smiljanic19A made their first contribution in https://github.com/livekit/agents/pull/7028 **Full Changelog**: https://github.com/livekit/agents/compare/livekit-agents@1.8.1...livekit-agents@1.8.2
Related Repositories
public-apis/public-apis★ 491,133
A collective list of free APIs
NousResearch/hermes-agent★ 255,957
The agent that grows with you
yt-dlp/yt-dlp★ 200,052
A feature-rich command-line audio/video downloader
microsoft/markitdown★ 191,490
Python tool for converting files and office documents to Markdown.
Significant-Gravitas/AutoGPT★ 187,703
AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.