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
test
语帆733073d2693d
diff --git a/funasr/auto/auto_model.py b/funasr/auto/auto_model.pyindex a3202fdb4..9db8c015d 100644--- a/funasr/auto/auto_model.py+++ b/funasr/auto/auto_model.py@@ -153,15 +153,18 @@ class AutoModel:# build tokenizertokenizer = kwargs.get("tokenizer", None)+ pdb.set_trace()if tokenizer is not None:tokenizer_class = tables.tokenizer_classes.get(tokenizer)+ pdb.set_trace()tokenizer = tokenizer_class(**kwargs["tokenizer_conf"])+ pdb.set_trace()kwargs["tokenizer"] = tokenizerkwargs["token_list"] = tokenizer.token_listvocab_size = len(tokenizer.token_list)else:vocab_size = -1-+ pdb.set_trace()# build frontendfrontend = kwargs.get("frontend", None)if frontend is not None: