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(server): return plain text for transcription text format (#3874)
be264e9 zhifu gao committed 8 days ago
fix(runtime): release ONNX non-WAV audio loading as 0.4.10 (#3876)
2b4a062 zhifu gao committed 8 days ago
Merge pull request #3866 from GPT-AGI/fix/timestamp-latin-leading-space
acbdbfe zhifu gao committed 8 days ago
Merge pull request #3863 from GPT-AGI/fix/python312-distutils
bb9de15 zhifu gao committed 8 days ago
fix(utils): preserve normalized boolean errors and cover Python 3.12 import
610aea3 zhifu gao committed 8 days ago
test(timestamp): align VAD sentence expectation with Latin spacing fix
e7e633a zhifu gao committed 8 days ago
fix(postprocess): keep a space between Latin VAD segments in rich transcription
c660e19 David-Wu1119 committed 8 days ago
fix(runtime): trim batch padding from CIF peaks before Paraformer timestamps
9396b7d Tanishq committed 8 days ago
fix(runtime): merge spelled-out acronyms into one word in sentence_postprocess
1c217c5 Tanishq committed 9 days ago
Merge pull request #3865 from tanishqbaweja/fix/onnx-non-wav-paths
7651777 zhifu gao committed 9 days ago
fix(onnx): load non-.wav audio paths with librosa instead of pydub
a3b21c4 Tanishq committed 10 days ago
Merge pull request #3857: verified regression coverage for #3854
b5d0fd9 zhifu gao committed 12 days ago
Merge pull request #3856: verified regression coverage for #3853
1d8979d zhifu gao committed 12 days ago
test: retain tokenizer regressions alongside normalization fixes
a3050a3 zhifu gao committed 12 days ago
Merge pull request #3855: verified regression coverage for #3852
d216ab8 zhifu gao committed 12 days ago
test(normalize): enforce precision and hosted regression coverage
1ad7430 zhifu gao committed 12 days ago
fix(paraformer): handle trailing zero-fire samples in mixed CIF batches
39e7511 BillVan1945 committed 12 days ago
fix(normalize): correct UtteranceMVN variance normalization
290c133 BillVan1945 committed 12 days ago
fix(tokenizer): ignore empty non-linguistic symbols to avoid a hang
5156f09 Mohammad Hijjawi committed 12 days ago