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
docs: add Training page with complete fine-tuning guide (data prep, multi-GPU, DeepSpeed, monitoring)
abc132e 游雁 committed 6 months ago
ci: add GitHub Actions workflow for auto-updating API docs on code changes
abffaa5 游雁 committed 6 months ago
docs: add detailed docstrings to ct_transformer_streaming, e_paraformer, fsmn_kws_mt, scama, llm_asr (class + inference methods)
dac02a0 游雁 committed 6 months ago
docs: add detailed method docstrings to emotion2vec, eres2net, bicif_paraformer
44dc655 游雁 committed 6 months ago
docs: add docstrings to 1950 functions across 276 files for API documentation
78cc132 游雁 committed 6 months ago
docs: add comprehensive docstrings to all AutoModel public methods
8e9bac7 游雁 committed 6 months ago
feat: add Qwen3-ASR speaker diarization; remove unnecessary punc_model from SPK examples
4967c8b 游雁 committed 6 months ago
docs: fix tutorial - len() syntax, add Fun-ASR-Nano section, remove duplicates
ab4f610 游雁 committed 6 months ago
feat: add ERes2NetV2 speaker model with full AutoModel integration
106cbb6 游雁 committed 6 months ago
refactor: Qwen3ASR follows standard AutoModel hub download convention
06e03d7 游雁 committed 6 months ago
fix: Fun-ASR-Nano batch check uses actual data size instead of kwargs batch_size
89ef5f7 游雁 committed 6 months ago