Repository Issue Activity (beta)

#377 Contribute to support new model: Nvidia Canny-qwen-1b

#371 Feature: Add FunASR as alternative simultaneous STT backend

#370 feat: Add SenseVoice/FunASR as alternative ASR backend — 5x faster, non-autoregressive

#369 Feature Request: Add SenseVoice/Paraformer as alternative STT backend

#368 Feature request: Add FunASR/SenseVoice as an ASR backend

#367 Feature: Add FunASR/SenseVoice as ASR backend (13x faster than Whisper)

#374 Timeout 120s hardcoded limits additional usecases

#347 前端没有文字。后端输出0.00s

#361 whisperlivekit config NLLB/NLLW translation zh ERROR

#373 Radeon compatibility

#372 Early Lines Missing in mode=full

#360 新版本我用的qwen3-asr-1.7b，好像不转写了

#277 Question : Is per-session language configuration possible? (multi-engine architecture)

#125 Feature Request: Pre-built Docker image

#135 I am new to python, I hope you can officially add the Chinese punctuation recovery function in the next version.

#349 Segmentation, diarization failing on clips longer than ~1min

#160 Feature Request: Add language parameter to AudioProcessor pipeline for transcription

#264 Translating once the transcription of the segment is complete.

#335 question mb add support SpeakerChangeDetector

#362 API /v1/* is not found

#358 load_cif: no CIF checkpoint → CIF is silently disabled (always_fire=True), cif_linear is dead code; no real word-boundary detection by default

#359 DecoderState.clean_cache() calls torch.cuda.empty_cache() on every inference step — hurts streaming latency

#365 Cumulative timestamp drift due to VAD delay double-counting in audio_processor.py

#363 VAD swallows short silence boundaries causing utterances to be incorrectly concatenated

#366 Feature suggestion: FunASR backend for lower streaming latency