Replies: 6 comments 4 replies
-
I'm also experiencing this problem, if there is a solution, please @ me, thank you very much! |
Beta Was this translation helpful? Give feedback.
-
Project member told me to Use ..\TTS\recipes\ljspeech\xtts_v2\train_gpt_xtts.py. Maybe it can work. You can finetune firstly. About infer, you should make some modification. |
Beta Was this translation helpful? Give feedback.
-
不幸的是,我用的就是这个脚本,hhhh, |
Beta Was this translation helpful? Give feedback.
-
你的报错是什么 |
Beta Was this translation helpful? Give feedback.
-
检查一下配置信息,以及dataset的格式。 check config and data format.config_dataset = BaseDatasetConfig( |
Beta Was this translation helpful? Give feedback.
-
the dir wavs remove and try again. only ../noah |
Beta Was this translation helpful? Give feedback.
-
Whether this problem have been fixed which has happened but fixed because of not supported xtts finetuning.
Traceback (most recent call last):
File "D:\Projects\TTS\TTS\bin\train_tts.py", line 71, in
main()
File "D:\Projects\TTS\TTS\bin\train_tts.py", line 58, in main
trainer = Trainer(
File "D:\Projects\TTS\venv\lib\site-packages\trainer\trainer.py", line 502, in init
self.criterion = self.get_criterion(self.model)
File "D:\Projects\TTS\venv\lib\site-packages\trainer\trainer.py", line 2077, in get_criterion
criterion = model.get_criterion()
File "D:\Projects\TTS\venv\lib\site-packages\torch\nn\modules\module.py", line 1695, in getattr
raise AttributeError(f"'{type(self).name}' object has no attribute '{name}'")
AttributeError: 'Xtts' object has no attribute 'get_criterion'
Beta Was this translation helpful? Give feedback.
All reactions