Unable to start XTTS training #74
Description
When trying to run a training session with the specified parameters (screenshot below), the program gives the error “Training failed”. Changing the audio file does not help. The console displays the following error:
2025-03-02 14:44:23,268 [INFO] XTTS Training: from TTS.tts.layers.xtts.trainer.gpt_trainer import GPTArgs, GPTTrainer, GPTTrainerConfig, XttsAudioConfig
2025-03-02 14:44:23 | INFO | root | XTTS Training: ImportError: cannot import name 'XttsAudioConfig' from 'TTS.tts.layers.xtts.trainer.gpt_trainer' (F:\Pandrator\Pandrator\conda\envs\easy_xtts_trainer\lib\site-packages\TTS\tts\layers\xtts\trainer\gpt_trainer.py)
2025-03-02 14:44:23,269 [INFO] XTTS Training: ImportError: cannot import name 'XttsAudioConfig' from 'TTS.tts.layers.xtts.trainer.gpt_trainer' (F:\Pandrator\Pandrator\conda\envs\easy_xtts_trainer\lib\site-packages\TTS\tts\layers\xtts\trainer\gpt_trainer.py)
2025-03-02 14:44:24 | INFO | root | XTTS Training: ERROR conda.cli.main_run:execute(125): `conda run python easy_xtts_trainer.py -i F:/Downloads/zer1.wav --source-language ru --whisper-model large-v3 --session Zerr --epochs 6 --gradient 1 --batch 2 --sample-method mixed --sample-rate 22050 --max-audio-time 11.0 --max-text-length 200 --method-proportion 6_4 --training-proportion 9_1 --denoise` failed. (See above for error)
2025-03-02 14:44:24,024 [INFO] XTTS Training: ERROR conda.cli.main_run:execute(125): `conda run python easy_xtts_trainer.py -i F:/Downloads/zer1.wav --source-language ru --whisper-model large-v3 --session Zerr --epochs 6 --gradient 1 --batch 2 --sample-method mixed --sample-rate 22050 --max-audio-time 11.0 --max-text-length 200 --method-proportion 6_4 --training-proportion 9_1 --denoise` failed. (See above for error).
