A FunASR backend, and diarization unbroken for qwen3.
Added
- FunASR SenseVoiceSmall backend (
--backend funasr): auto, Mandarin, Cantonese, English, Japanese and Korean on the LocalAgreement policy, with fail-closed timestamp validation and per-session language overrides. Install with thefunasrextra. (#385, thanks @LauraGPT) - Published Docker images now carry SBOM attestations, and the provenance mode is pinned explicitly. (#390, thanks @kobihikri)
Fixed
- qwen3 backends with
--diarizationsilently stopped emitting updates as soon as the first line was diarized: the qwen3 package's own token type lacked the helpers the alignment needs. Tokens are now normalized to WhisperLiveKit'sASRTokenat the factory boundary. (#389, thanks @OookTheLibrarian) - Diarization span corruption in
concatenate_diar_segments. --language yue --buffer-trimming sentencecrashed on a bare assert even though the sentence tokenizer supports Cantonese. (#392)- Invalid per-session parameters (such as an unsupported
?language=) now produce a structured error and close code 4400 on the WebSocket endpoints, and HTTP 400 on the REST endpoint, instead of an unexplained handshake failure. (#393) wlk pull funasrandwlk rm funasrexplain that funasr manages its own model cache instead of exiting 1 silently. (#394)