search

jaywalnut310 / vits

VITS: Conditional Variational Autoencoder with Adversarial Learning for End-to-End Text-to-Speech
https://jaywalnut310.github.io/vits-demo/index.html
MIT License
6.72k stars 1.23k forks source link

Is there a way to do batch inference? #190

Open SoulflareRC opened 11 months ago

SoulflareRC commented 11 months ago

I'm trying to do inference in batch by padding each sentence to the max length(pad by 0), but it turns out that the model did fine on the text part, but generated noises for the padded part. Is there a proper way to do this?

sushant-samespace commented 9 months ago

Even I am looking for a solution to this. Were you able to find anything ?

piyushsinghpasi commented 8 months ago

I created batched inference script since I had to run things on scale. Feel free to use it: https://github.com/piyushsinghpasi/VITS-Batched-Inference

BB-Beta commented 6 months ago

I created batched inference script since I had to run things on scale. Feel free to use it: https://github.com/piyushsinghpasi/VITS-Batched-Inference

thanks，I have the same problem and solved by your solution