livekit-agents@1.5.14
hpcaitech/Open-Soralivekit-agents@1.5.14May 27, 2026by tinalenguyen
AI Summary
Focuses on voice and avatar enhancements, adding new STT plugins, models, and visual features like background audio fades.
Key Highlights
- Added gpt-realtime-2 model string
- Added Inworld TTS delivery_mode parameter
- Added GnaniAI STT plugin
- Added background audio fade_in/fade_out to AudioConfig
- Added LemonSlice avatar with switchable personas
New Features
- gpt-realtime-2 model
- Inworld TTS delivery_mode
- GnaniAI STT plugin
- Audio fade effects
- LemonSlice avatar
Full Release Notes
## What's Changed * (openai realtime): add gpt-realtime-2 model str by @tinalenguyen in https://github.com/livekit/agents/pull/5791 * (healthcare example): remove filesearch and pdf by @tinalenguyen in https://github.com/livekit/agents/pull/5795 * chore(worker): update worker warnings AGT-2909 by @chenghao-mou in https://github.com/livekit/agents/pull/5771 * fix(voice): race between flush and clear_buffer on interrupt leaks unplayed transcript by @longcw in https://github.com/livekit/agents/pull/5798 * fix(vad): add support for vad reset directly without stream close by @chenghao-mou in https://github.com/livekit/agents/pull/5687 * feat(inworld tts): add delivery_mode parameter for inworld-tts-2 by @chrisackermann in https://github.com/livekit/agents/pull/5801 * fix(voice): wait for end-of-turn task when waiting for user inactivity by @longcw in https://github.com/livekit/agents/pull/5792 * Adding GnaniAI STT plugin by @Gnani-AI-Mintlify in https://github.com/livekit/agents/pull/5769 * (gnani): fix py.typed and remove changesets by @tinalenguyen in https://github.com/livekit/agents/pull/5816 * fix(voice): propagate ChatMessage.interrupted through proto serializer by @toubatbrian in https://github.com/livekit/agents/pull/5824 * fix(anthropic): recreate stream on retry by @he-yufeng in https://github.com/livekit/agents/pull/5820 * fix(llm): convert per-turn instructions on the very first turn too by @theomonnom in https://github.com/livekit/agents/pull/5828 * feat(background_audio): add fade_in / fade_out to AudioConfig by @theomonnom in https://github.com/livekit/agents/pull/5832 * fix(plugins/sarvam): thread language_probability into SpeechData.confidence by @hashirventhodi in https://github.com/livekit/agents/pull/5830 * fix(lemonslice): bind avatar audio output before the upstream session HTTP call by @theomonnom in https://github.com/livekit/agents/pull/5837 * feat(voice/avatar): kick avatar participant on aclose + wait_for_join helper by @theomonnom in https://github.com/livekit/agents/pull/5836 * feat(examples): add LemonSlice avatar with switchable personas by @theomonnom in https://github.com/livekit/agents/pull/5834 * chore(playground): list frontdesk first by @theomonnom in https://github.com/livekit/agents/pull/5838 * fix: 4xx errors should not be retryable by @davidzhao in https://github.com/livekit/agents/pull/5831 * livekit-agents@1.5.13 by @github-actions[bot] in https://github.com/livekit/agents/pull/5835 * feat(assemblyai): add continuous_partials and interruption_delay stre… by @dlange-aai in https://github.com/livekit/agents/pull/5819 * ci(deploy-examples): include avatar in matrix + wire LEMONSLICE_API_KEY by @theomonnom in https://github.com/livekit/agents/pull/5839 * docs: clarify MCP support bullet by @scosemicolon in https://github.com/livekit/agents/pull/5822 * fix(core): replace frame drop with silence frame AGT-2914 by @chenghao-mou in https://github.com/livekit/agents/pull/5815 * feat(plugins-soniox): surface per-run language segments end-to-end by @MSameerAbbas in https://github.com/livekit/agents/pull/5730 * use stt timestamps as last speaking time by @chenghao-mou in https://github.com/livekit/agents/pull/5672 * docs: fix Gemini model version in avatar example README by @detail-app[bot] in https://github.com/livekit/agents/pull/5842 * handle leaked chat-template tokens in function call args by @davidzhao in https://github.com/livekit/agents/pull/5840 * fix(voice): restrict stt pipeline reuse to default stt_node by @longcw in https://github.com/livekit/agents/pull/5803 * feat(google): integration for AI Platform LLMs by @davidzhao in https://github.com/livekit/agents/pull/5843 * fix: return function argument errors (ToolError) to LLM by @davidzhao in https://github.com/livekit/agents/pull/5846 * refactor(llm): unified ToolError contract for tool arg validation by @longcw in https://github.com/livekit/agents/pull/5807 * feat(openai): stream input_audio_transcription delta events by @longcw in https://github.com/livekit/agents/pull/5859 * fix(voice): block on_user_turn_exceeded during agent handoff by @longcw in https://github.com/livekit/agents/pull/5858 * fix(voice): reset user turn tracker on clear_user_turn by @longcw in https://github.com/livekit/agents/pull/5857 * fix(elevenlabs): always close TTS stream context on cancellation by @longcw in https://github.com/livekit/agents/pull/5845 * Update download-files deprecation message by @bcherry in https://github.com/livekit/agents/pull/5781 * fix(llm): preserve Field() constraints on function tool arguments by @theomonnom in https://github.com/livekit/agents/pull/5861 * fix(soniox): surface STT server errors by @he-yufeng in https://github.com/livekit/agents/pull/5864 * Fix ElevenLabs websocket context id handling by @flynn-hamming in https://github.com/livekit/agents/pull/5813 * livekit-agents@1.5.14 by @github-actions[bot] in https://github.com/livekit/agents/pull/5870 ## New Contributors * @Gnani-AI-Mintlify made their first contribution in https://github.com/livekit/agents/pull/5769 * @he-yufeng made their first contribution in https://github.com/livekit/agents/pull/5820 * @hashirventhodi made their first contribution in https://github.com/livekit/agents/pull/5830 * @scosemicolon made their first contribution in https://github.com/livekit/agents/pull/5822 * @flynn-hamming made their first contribution in https://github.com/livekit/agents/pull/5813 **Full Changelog**: https://github.com/livekit/agents/compare/livekit-agents@1.5.12...livekit-agents@1.5.14