What's Changed
- fix(core): reset user away timer on final STT transcript AGT-3149 by @chenghao-mou in #6478
- Fix stripping of Gemma reasoning markers in streamed output by @C0deRatoR in #6398
- feat: Spatius avatar plugin by @3DRX in #5821
- fix(avatar): skip removal if avatar not found in the room by @longcw in #6486
- fix(endpointing): keep max_delay constant in dynamic endpointing by @chenghao-mou in #6265
- fix(stt): don't clear the primary speaker on an unattributed segment by @chuenchen309 in #6460
- fix(drive-thru): assume quantity of one when unstated to avoid count confirmation by @u9g in #6463
- fix: make AMD settle on endpointing backstop by default by @chenghao-mou in #6496
- fix(voice): include function name in FunctionCallOutput protos by @u9g in #6491
- feat: tag simulation and redaction telemetry by @chenghao-mou in #6497
- Update READMEs to use gemma by @keepingitneil in #6502
- fix(voice): don't drop the realtime turn when the chat-ctx push times out by @ByteMaster-1 in #6445
- feat(voice): expose realtime provider request IDs by @yaodong-shen in #6501
- Add 'zh-HK' to DeepgramLanguages by @devb-enp in #6507
- (gnani): Update authors in pyproject.toml by @tinalenguyen in #6489
- Allow local VAD inference to use a dedicated executor by @basnijholt in #6498
- feat(fishaudio): add normalize_loudness and generation-tuning TTS options by @adrian-cowham in #6520
- fix(voice): don't hold session.run() open on update_agent handoffs with long-lived on_enter by @k-zaher in #6500
- plugins(krisp): add voice isolation telephony mode by @lukasIO in #6510
- Support adaptive interruption for realtime models by @longcw in #6488
- feat(inference): add inference.AvatarSession (avatar provisioning via the gateway) by @shawnfeldman in #6492
- (inference): add fish audio tts by @tinalenguyen in #6529
- slng: rewrite plugin around the Unmute Bridge (v2) by @metehan-slng in #6442
- feat(speechify): streaming TTS with word-level timestamps via official SDK by @luke-speechify in #6327
- fix(openai): recover from stale pooled Responses WebSocket connections by @longcw in #6523
- prewarm LLM to skip connection establishment by @davidzhao in #6484
- fix(otel): skip otel upload error logs if it's disabled by the project setting AGT-3015 by @chenghao-mou in #6167
- chore(spatius): bump spatius to 1.0.4 by @3DRX in #6536
- fix(livekit-agents): require livekit-protocol>=1.1.21 for MetricsRecordingHeader.simulated by @u9g in #6541
- (xai tts): add new voices by @tinalenguyen in #6543
- fix(speechify): restore caller header, close owned httpx client by @luke-speechify in #6542
- fix(plugins-aws): Polly TTS silent on aiobotocore >=3.8.0 (AioStreamingBody has no .content) by @Gooh456 in #6540
- feat(telemetry): include agent_name as an OTLP resource attribute by @jmcclanahan in #6539
- livekit-agents@1.6.7 by @github-actions[bot] in #6532
New Contributors
- @C0deRatoR made their first contribution in #6398
- @3DRX made their first contribution in #5821
- @chuenchen309 made their first contribution in #6460
- @yaodong-shen made their first contribution in #6501
- @basnijholt made their first contribution in #6498
- @k-zaher made their first contribution in #6500
- @shawnfeldman made their first contribution in #6492
- @luke-speechify made their first contribution in #6327
- @Gooh456 made their first contribution in #6540
- @jmcclanahan made their first contribution in #6539
Full Changelog: https://github.com/livekit/agents/compare/livekit-agents@1.6.6...livekit-agents@1.6.7