how can I change voice?

[hansolGib]

this is really interesting TTS. I'm trying to change the voice, but how can I achieve it? I tried to record my voice reading BASIC5000_0001 and 0002, and I put them in jp_dataset/basic5000/wav folder. and then I tried to run python3 train.py -c configs/jp_base.json -m jp_base this command in my Macbook but I got an error File "/Users/hansol/Downloads/vits-japanese-main/train.py", line 43, in main assert torch.cuda.is_available(), "CPU training is not allowed." I guess it's because I'm using MacBook?. and I would like to confirm that this is the correct way to change the voice of TTS.

[QuyAnh2005]

@hansolGib Error File "/Users/hansol/Downloads/vits-japanese-main/train.py", line 43, in main assert torch.cuda.is_available(), "CPU training is not allowed." occured because the training process is required to use GPU (not use CPU - You are using CPU for training) Morever, if you are trying to change the voice by yourself, you need a lot of audios to train (often more 2 hours). An alternative way is that you should use TTS, then use a Voice Conversion system such as:

• FragmentVC - not realtime
• styleTTS-VC (I will upload in the future) - realtime

Hope that it helps you.

[QuyAnh2005]

@hansolGib StyleTTS-VC for Japanese is available now at here