Open ParthThakkar08 opened 1 year ago
Yes I have the same problem as the output is changed so much different with the input in tune & style
Hi, ParthThakkar08 and universeroc. I am also try to train the model for hindi text and voice. i am getting error under train_semantic code in setuping the datasets for the model. If you guy's have something to share please share it. Hoping positive response from your side.
Do we have any options for generating a voice actually matching the input file?
I am trying to clone the voice for hindi but it changes the voice very much so I wanted to know if we can anyhow create a new model or upgrade the voice we have in model or something? I am also happy to create a new model for it and we can work on this together if you guys don't mind.
Thanks