Version Packages#2029
Merged
Merged
Conversation
github-actions
Bot
force-pushed
the
changeset-release/main
branch
from
July 14, 2026 04:43
b885841 to
a74de77
Compare
![devin-ai-integration[bot]](https://avatars.githubusercontent.com/in/811515?s=60&v=4){kind=small}
github-actions
Bot
force-pushed
the
changeset-release/main
branch
19 times, most recently
from
July 18, 2026 00:01
a50c825 to
4372d42
Compare
github-actions
Bot
force-pushed
the
changeset-release/main
branch
from
July 18, 2026 00:02
4372d42 to
64aeb41
Compare
toubatbrian
approved these changes
Jul 18, 2026
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR was opened by the Changesets release GitHub action. When you're ready to do a release, you can merge this and the packages will be published to npm automatically. If you're not ready to do a release yet, that's fine, whenever you add more changesets to main, this PR will be updated.
Releases
@livekit/agents@1.5.3
Patch Changes
Export the
AudioInputbase class from thevoicemodule so plugins and applications can implement custom agent audio inputs. - #2036 (@aweitz)Commit the in-flight assistant turn (interrupted: true, partially-forwarded text) when the session closes mid-playout — previously a room disconnect during playback dropped the turn from chatCtx entirely, with no ConversationItemAdded emitted (Assistant turn lost entirely (no ConversationItemAdded, nothing in chatCtx) when the room disconnects mid-playout #2041) - #2042 (@Sarfaraz85)
Restart chunked TTS retries with a fresh attempt queue so retried synthesis uses a fresh request ID and does not write to a closed failed-attempt queue. - #1994 (@rosetta-livekit-bot)
Honor
ToolFlag.IGNORE_ON_ENTERfor tools nested insideToolsets. - #2011 (@rosetta-livekit-bot)Fix modality-aware instruction templates to collapse identical variants and avoid duplicate rendered output. - #2030 (@rosetta-livekit-bot)
fix(voice): stop dropping agent turns whose playback starts after audio forwarding completes (Agent speech silently dropped when interrupted before the first audio frame (resumeFalseInterruption) — port of Python #5039 #1909, fix(voice): commit interrupted replies when playback start arrives after forwarding completes (avatar outputs) #1960; port of fix(core): move callbacks to the caller agents#5039). - #1966 (@toubatbrian)
forwardAudioused to rejectfirstFrameFut(and detach itsPLAYBACK_STARTEDlistener) in itsfinallyblock whenever no frame had played by the time forwarding finished. Two real scenarios hit this window: a speech paused in the thinking state by a brief user sound, whose buffered first frame only plays after the false interruption clears (Agent speech silently dropped when interrupted before the first audio frame (resumeFalseInterruption) — port of Python #5039 #1909), and DataStream avatar outputs withwaitPlaybackStart: true, which deliverlk.playback_started~1s after frames were captured (fix(voice): commit interrupted replies when playback start arrives after forwarding completes (avatar outputs) #1960). In both cases the late playback-started event found nothing listening, the reply was classified "skipped", and the turn was silently removed from the chat context while the agent never entered thespeakingstate.The
PLAYBACK_STARTEDlistener now lives inperformAudioForwardingso it outlives the forwarding task, andforwardAudiono longer settles the future; the reply tasks (including thesay()path) settle it once the playout window ends, which also detaches the listener. A reported non-zero playback position on interruption is additionally honored as evidence of partial playback — but only when the segment actually captured a frame into the output (tracked via the output's segment count, which is also what makes the reported position fresh rather than stale) — covering avatars whose playback-started RPC races the interruption itself.Preserve partial LLM response telemetry and generated function calls on every inference exit path. - #2048 (@toubatbrian)
fix(voice): stop RecorderIO from dropping the final agent speech at session teardown. A force-interrupted shutdown marks the current speech done before playout settles, so the recorder could close and fence out the in-flight playbackFinished flush, silently losing the last agent turn and trailing mic audio from the recording. RecorderIO.close() now waits (bounded) for the pending playback event — which carries the authoritative playback position — before fencing, flushes any input captured since the last write, and warns if unflushed agent audio had to be dropped. - #2037 (@chenghao-mou)
Deprecate the
nativeTranscriptSyncrealtime model capability while preserving its existing transcript synchronization behavior for third-party models. Remove Phonic's redundant explicit opt-out now that it usesstream_ahead_of_real_timemode. - #2044 (@tinalenguyen)Avoid splitting streamed replacement output mid-word when no replacement key prefix is pending. - #2050 (@rosetta-livekit-bot)
Add a beta DTMF sending tool that publishes to the active agent session room. - #2010 (@rosetta-livekit-bot)
Add
speedto xAI TTS inference options. - #2028 (@rosetta-livekit-bot)@livekit/agents-plugin-anam@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-anthropic@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-assemblyai@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-azure@1.5.3
Patch Changes
Add Azure streaming STT with cancellation error recovery. - #2009 (@rosetta-livekit-bot)
Updated dependencies [
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-baseten@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-bey@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-cartesia@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-cerebras@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6a998f6,c0c927a,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-deepgram@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-did@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-elevenlabs@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-fishaudio@1.5.3
Patch Changes
Drop Fish Audio's startup audio prebuffer so streaming audio starts from the opening chunk. - #2033 (@rosetta-livekit-bot)
Reuse and prewarm Fish Audio streaming TTS websockets to avoid repeated connection setup. - #2017 (@rosetta-livekit-bot)
Expose Fish Audio TTS sampling, bitrate, and normalization options. - #2049 (@rosetta-livekit-bot)
Updated dependencies [
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-google@1.5.3
Patch Changes
Gate mixed Gemini built-in and function tool requests to the Gemini 3 Developer API and build Google tool config in chat requests. - #2035 (@rosetta-livekit-bot)
Strip Gemini thought summary parts from Google LLM text output. - #2039 (@rosetta-livekit-bot)
Updated dependencies [
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-hedra@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-hume@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-inworld@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-lemonslice@1.5.3
Patch Changes
Add the ability for users to provide a local image to the LemonSlice transport to be used as the avatar image. - #1986 (@jp-lemon)
Convert extrapayload to snake_case - #2034 (@aweitz)
Add external meeting support to the LemonSlice avatar plugin:
joinMeeting()andleaveMeeting()send the avatar into Zoom, Google Meet, Microsoft Teams, or Webex meetings, stream mixed meeting audio into the agent session's STT, and optionally relay meeting chat messages as user input. - #2036 (@aweitz)Updated dependencies [
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-liveavatar@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-livekit@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-minimax@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-mistral@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-mistralai@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-neuphonic@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-openai@1.5.3
Patch Changes
Stop OpenAI realtime sessions from reconnecting forever after fatal server errors. - #2031 (@rosetta-livekit-bot)
Include OpenAI model names on Responses WebSocket URLs for gateway endpoints. - #2032 (@rosetta-livekit-bot)
Updated dependencies [
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-perplexity@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6a998f6,c0c927a,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-phonic@1.5.3
Patch Changes
Deprecate the
nativeTranscriptSyncrealtime model capability while preserving its existing transcript synchronization behavior for third-party models. Remove Phonic's redundant explicit opt-out now that it usesstream_ahead_of_real_timemode. - #2044 (@tinalenguyen)Updated dependencies [
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-resemble@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-rime@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-runway@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-sarvam@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-silero@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-soniox@1.5.3
Patch Changes
Add Soniox TTS support with configurable speaking speed. - #2005 (@rosetta-livekit-bot)
Add Soniox endpoint latency adjustment level STT option. - #2027 (@rosetta-livekit-bot)
Updated dependencies [
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-tavus@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-trugen@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugin-xai@1.5.3
Patch Changes
Add xAI TTS support with websocket connection rotation before xAI's server deadline. - #2025 (@rosetta-livekit-bot)
Updated dependencies [
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6a998f6,c0c927a,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]:@livekit/agents-plugins-test@1.5.3
Patch Changes
3c19945,7db92dd,b9d8e95,bad25fd,206e884,16ac15e,6300c1c,dfd6654,97c5aa2,13a191c,508236a,2312eed]: