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
Update auto_model.py
liugz18bac3ef2e250c
diff --git a/funasr/auto/auto_model.py b/funasr/auto/auto_model.pyindex 9f8e4dc13..603c0a09d 100644--- a/funasr/auto/auto_model.py+++ b/funasr/auto/auto_model.py@@ -466,23 +466,21 @@ class AutoModel:result[k] = restored_data[j][k]else:result[k] += restored_data[j][k]-++ if not len(result["text"].strip()):+ continuereturn_raw_text = kwargs.get("return_raw_text", False)# step.3 compute punc modelraw_text = Noneif self.punc_model is not None:- if not len(result["text"].strip()):- if return_raw_text:- result["raw_text"] = raw_text = ""- else:- deep_update(self.punc_kwargs, cfg)- punc_res = self.inference(- result["text"], model=self.punc_model, kwargs=self.punc_kwargs, **cfg- )- raw_text = copy.copy(result["text"])- if return_raw_text:- result["raw_text"] = raw_text- result["text"] = punc_res[0]["text"]+ deep_update(self.punc_kwargs, cfg)+ punc_res = self.inference(+ result["text"], model=self.punc_model, kwargs=self.punc_kwargs, **cfg+ )+ raw_text = copy.copy(result["text"])+ if return_raw_text:+ result["raw_text"] = raw_text+ result["text"] = punc_res[0]["text"]# speaker embedding cluster after resortedif self.spk_model is not None and kwargs.get("return_spk_res", True):