search

p0p4k / vits2_pytorch

unofficial vits2-TTS implementation in pytorch
https://arxiv.org/abs/2307.16430
MIT License
477 stars 85 forks source link

colab error #73

Open alexxony opened 9 months ago

alexxony commented 9 months ago

I should have to do this for unsetting espeak !apt-get install espeak

Also It occured AttributeError: 'HParams' object has no attribute 'duration_discriminator_type'

p0p4k commented 9 months ago

configs/vits2_vctk_standard.json follow that file and change your json to add 'duration_discriminator_type'

alexxony commented 9 months ago

configs/vits2_vctk_standard.json follow that file and change your json to add 'duration_discriminator_type'

https://colab.research.google.com/drive/1y_XFXZTcWmV7l0w1anZqfD4zrEdXGwQ5?usp=sharing

other error occured. how can i solve that?

p0p4k commented 9 months ago

if you are talking about this error - torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate 20.00 MiB. GPU 0 has a total capacty of 14.75 GiB of which 5.06 MiB is free. Process 33947 has 14.74 GiB memory in use. Of the allocated memory 14.26 GiB is allocated by PyTorch, and 200.53 MiB is reserved by PyTorch but unallocated. If reserved but unallocated memory is large try setting max_split_size_mb to avoid fragmentation. See documentation for Memory Management and PYTORCH_CUDA_ALLOC_CONF it says that your gpu is out of memory, so reduce batch size number and try again. I will advice you to google search the errors to debug on your own first and read a bit more about training models. Thanks.