All pull requests

Pull requests

• fix(ipc): update the active-job gauge in the thread job executor

  #7920·BlueX888 opened Oct 18, 2026Contributor·

  ·
  3

• feat(voice): option to start a reply on time when the caller only made a noise

  #7919·Rehansanjay opened Oct 18, 2026Contributor·

  ·
  1

  2

• fix(inference): log retryable STT recv errors at debug

  #7918·muntazir868 opened Oct 18, 2026·

  ·
  6

• fix(stt): notice a half-open websocket in the gladia, nabrah, sarvam and slng streams

  #7914·Rehansanjay opened Oct 18, 2026Contributor·

  ·
  2

• fix(stt): notice a half-open websocket in the openai, elevenlabs and inworld streams

  #7913·Rehansanjay opened Oct 18, 2026Contributor·

  ·
  1

  4

• fix(stt): detect half-open websockets in OpenAI, ElevenLabs and Inworld STT

  #7912·ninglinLiu opened Oct 17, 2026·

  ·
  1

  1

• feat(avaz): add Avaz Turkish TTS plugin

  #7911·tamerrkanak opened Oct 17, 2026·

  ·
  7

• fix(voice): record a realtime text turn's message before pushing it

  #7910·Rehansanjay opened Oct 17, 2026Contributor·

  ·
  1

  4

• fix(anthropic): keep thinking blocks across tool calls

  #7909·Yatha04 opened Oct 17, 2026·

  ·
  1

  7

• fix(realtime): keep the reply in progress when the session reconnects

  #7907·Rehansanjay opened Oct 17, 2026Contributor·

  ·
  3

• fix(voice): preserve tool outputs across concurrent realtime context updates

  #7905·Epsilon006 opened Oct 17, 2026·

  ·
  1

  6

• fix(voice): vectorize the speaking-rate spectral flux

  #7903·omerekin opened Oct 17, 2026·

  ·
  1

  1

• fix(ipc): unblock thread executor aclose() when initialize is cancelled

  #7899·BlueX888 opened Oct 17, 2026Contributor·

  ·
  9

• fix: reuse in-hand final transcript in commit_user_turn (manual turns)

  #7898·iamdgarcia opened Oct 17, 2026·

  ·
  1

  5

• fix(soniox): emit END_OF_SPEECH when an endpoint follows a fully retracted segment

  #7897·iamdgarcia opened Oct 17, 2026·

  ·
  1

• fix(google): report realtime usage metrics once per generation

  #7896·BlueX888 opened Oct 17, 2026Contributor·

  ·
  6

• fix(voice): resolve the aligned-transcript wait when TTS fails before first yield

  #7895·BlueX888 opened Oct 17, 2026Contributor·

  ·

• fix(ipc): fail pending inference requests when the inference process dies

  #7893·BlueX888 opened Oct 17, 2026Contributor·

  ·
  6

• fix(cli): exit non-zero when the worker fails to start

  #7892·BlueX888 opened Oct 17, 2026Contributor·

  ·
  3

• fix(google): count Gemini thinking tokens in LLM completion_tokens

  #7891·ninglinLiu opened Oct 17, 2026·

  ·
  1

  1

• fix(meta): don't delay Muse STT audio by the websocket connect time

  #7890·DEVJHAWAR11 opened Oct 17, 2026·

  ·
  1

  7

• fix(anthropic): don't add message_delta output tokens to the message_start count

  #7888·pcbeingused333 opened Oct 17, 2026·

  ·
  1

• fix(inference): emit every parallel tool call sent in one chunk

  #7887·pcbeingused333 opened Oct 17, 2026·

  ·
  1

• fix(boson-avatar): follow the Boson API's paths and object names

  #7886·warlockee opened Oct 17, 2026Contributor·

  ·
  3

• feat(azure): use the Azure Speech SDK for TTS

  #7884·tinalenguyen opened Oct 16, 2026Member·

  ·
  3