Repository Issue Activity (beta)

#413 Implement correct Deepgram endpointing and utterance semantics

#410 Support a tested max_speakers limit for Sortformer diarization

#409 Add configurable pause-based transcript segmentation

#335 Split transcript segments on detected speaker changes

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

#386 REST: diarized response format, usage.duration and real word timestamps

#397 Code

#373 Radeon compatibility

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

#347 前端没有文字。后端输出0.00s

#394 wlk pull/rm funasr exits 1 with no output

#393 Unsupported per-session language should close the session cleanly instead of raising before accept()

#392 funasr: --language yue with --buffer-trimming sentence crashes on a bare assert

#388 Compute time displayed in localhost:8000 window is about 40 percent of elapsed time

#366 Feature suggestion: FunASR backend for lower streaming latency

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

#372 Early Lines Missing in mode=full

#362 API /v1/* is not found

#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.

#360 新版本我用的qwen3-asr-1.7b，好像不转写了

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

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

#374 Timeout 120s hardcoded limits additional usecases

#371 Feature: Add FunASR as alternative simultaneous STT backend