kumdori88

Thank you for the reply! I got these errors despite changing from init.py and train.py to TextMelSpeakerLoader, TextMelSpeakerCollate. I also attach the code for your reference. **===init,py===** ``` import argparse...

Yes I am still getting the error. Is there any code I can refer to? Or maybe it's a CUDA or GPU issue?

Thank you so much for your kind reply. After making some modifications, a new error occurred. ``` INFO:korean_base_ms:{'train': {'use_cuda': True, 'log_interval': 20, 'seed': 1234, 'epochs': 10000, 'learning_rate': 1.0, 'betas': [0.9,...

I understand now about tensor shapes. The number of speakers is 12, and starting with number 1. Below is the speaker data I have configured. ``` DUMMY_KSS/1/1_0000.wav|1|그는 괜찮은 척하려고 애쓰는...

Thank you so much. Learning goes on! I never even thought that the speaker ID would start with 0.