Open-source speech recognition toolkit for training, inference, streaming ASR, VAD, punctuation, speaker diarization pipelines, and OpenAI-compatible/MCP serving.
Python asraudiochineseemotion-recognitionfunasrmcp-servermultilingual-asropenai-compatible-apiparaformerpunctuationpytorchreal-time-asrspeaker-diarizationspeech-recognitionspeech-to-textstreaming-asrtranscriptionvllmvoice-activity-detectionwhisper-alternative
Commits on main
fix(onnxruntime): Fix Windows MSVC compilation errors (#2841)
6902805 imdenny committed 6 months ago
fix(runtime/go): websocket client premature close in offline mode (#2834)
13b244d 游雁 committed 6 months ago
fix: propagate hub parameter to vad/punc/spk sub-model builds (#2859)
364dc97 游雁 committed 6 months ago
Revert "readme: replace SVG banner with value-first header (code demo + badges + one-liner)"
c8ebbc7 游雁 committed 6 months ago
readme: replace SVG banner with value-first header (code demo + badges + one-liner)
6f0778d 游雁 committed 6 months ago
fix: increase inference_batch_size from 1 to 32 in aishell recipes
0cbf26e 游雁 committed 6 months ago
benchmark: remove GLM-ASR (community model, speed needs verification)
e63d235 游雁 committed 6 months ago
benchmark: expand to 8 GPU models + 3 CPU models with full comparison
18344cd 游雁 committed 6 months ago
Add benchmark section: FunASR 15x faster than Whisper with lower CER
6122ec1 游雁 committed 6 months ago
fix: improve streaming vLLM with two-stage approach for long audio
4bb5b54 游雁 committed 6 months ago
fix: skip punc_model for models with native punctuation output (Fun-ASR-Nano, SenseVoice, Qwen3-ASR)
7a70f18 游雁 committed 6 months ago