FunAudioLLM / SenseVoice

Multilingual Voice Understanding Model
https://funaudiollm.github.io/
Other
3.35k stars 307 forks source link

could not parse ModelProto from iic/SenseVoiceSmall/chn_jpn_yue_eng_ko_spectok.bpe.model #22

Closed Rasin-wu closed 4 months ago

Rasin-wu commented 4 months ago

File "run.py", line 32, in model = AutoModel(model=model_dir, File "/src/funasr/FunASR/funasr/auto/auto_model.py", line 124, in init model, kwargs = self.build_model(kwargs) File "/src/funasr/FunASR/funasr/auto/auto_model.py", line 192, in build_model tokenizer = tokenizer_class(kwargs.get("tokenizer_conf", {})) File "/src/funasr/FunASR/funasr/tokenizer/sentencepiece_tokenizer.py", line 23, in init self._build_sentence_piece_processor() File "/src/funasr/FunASR/funasr/tokenizer/sentencepiece_tokenizer.py", line 32, in _build_sentence_piece_processor self.sp.load(self.bpemodel) File "/tools/anaconda3/envs/funasr/lib/python3.8/site-packages/sentencepiece/init.py", line 961, in Load return self.LoadFromFile(model_file) File "/tools/anaconda3/envs/funasr/lib/python3.8/site-packages/sentencepiece/init.py", line 316, in LoadFromFile return _sentencepiece.SentencePieceProcessor_LoadFromFile(self, arg) RuntimeError: Internal: could not parse ModelProto from iic/SenseVoiceSmall/chn_jpn_yue_eng_ko_spectok.bpe.model

请问我用的最新的funasr-1.1.0, 也是最新下的iic/SenseVoiceSmall 出现这个bpe model的解析错误会是什么原因呢

Rasin-wu commented 4 months ago

原因已经排查,在没有检查git lfs的情况下直接git clone SenseVoiceSmall模型,导致出错