使用bert_vits2的模型推理时报错

[v3ucn]

File "E:\work\vits-simple-api\bert_vits2\text\chinese_bert.py", line 35, in get_bert_feature res = model(**inputs, output_hidden_states=True) NameError: name 'model' is not defined

[Artrajz]

检查bert模型下了吗

[v3ucn]

刚刚下好了，但是开始报这个错了：

D:\work\vits-simple-api>python app.py DEBUG:tzlocal:Looking up time zone info from registry INFO:apscheduler.scheduler:Scheduler started INFO:vits-simple-api:Loading chinese-roberta-wwm-ext-large... Some weights of the model checkpoint at D:\work\vits-simple-api/bert_vits2/bert/chinese-roberta-wwm-ext-large were not used when initializing BertForMaskedLM: ['bert.pooler.dense.weight', 'bert.pooler.dense.bias', 'cls.seq_relationship.weight', 'cls.seq_relationship.bias']

• This IS expected if you are initializing BertForMaskedLM from the checkpoint of a model trained on another task or with another architecture (e.g. initializing a BertForSequenceClassification model from a BertForPreTraining model).
• This IS NOT expected if you are initializing BertForMaskedLM from the checkpoint of a model that you expect to be exactly identical (initializing a BertForSequenceClassification model from a BertForSequenceClassification model). INFO:vits-simple-api:Loading finished. INFO:vits-simple-api:Loading bert-base-japanese-v3... ERROR:vits-simple-api:You need to install fugashi to use MecabTokenizer. See https://pypi.org/project/fugashi/ for installation. ERROR:vits-simple-api:Please download pytorch_model.bin from cl-tohoku/bert-base-japanese-v3. Building prefix dict from the default dictionary ... DEBUG:jieba:Building prefix dict from the default dictionary ... Loading model from cache C:\Users\zcxey\AppData\Local\Temp\jieba.cache DEBUG:jieba:Loading model from cache C:\Users\zcxey\AppData\Local\Temp\jieba.cache Loading model cost 0.331 seconds. DEBUG:jieba:Loading model cost 0.331 seconds. Prefix dict has been built successfully. DEBUG:jieba:Prefix dict has been built successfully. ERROR:vits-simple-api:Uncaught exception Traceback (most recent call last): File "D:\work\vits-simple-api\app.py", line 30, in tts = load_model(app.config["MODEL_LIST"]) File "D:\work\vits-simple-api\utils\load_model.py", line 172, in load_model from bert_vits2 import Bert_VITS2 File "D:\work\vits-simple-api\bert_vits2__init__.py", line 1, in from bert_vits2.bert_vits2 import Bert_VITS2 File "D:\work\vits-simple-api\bert_vits2\bert_vits2.py", line 8, in from bert_vits2.text.cleaner import clean_text File "D:\work\vits-simple-api\bert_vits2\text\cleaner.py", line 1, in from bert_vits2.text import chinese, japanese, cleaned_text_to_sequence File "D:\work\vits-simple-api\bert_vits2\text\japanese.py", line 9, in from bert_vits2.text.japanese_bert import tokenizer ImportError: cannot import name 'tokenizer' from 'bert_vits2.text.japanese_bert' (D:\work\vits-simple-api\bert_vits2\text\japanese_bert.py)

[v3ucn]

可以了，貌似时1.1的模型和1.0模型不通用导致的，感谢。