-
|
After using this project for training, the timbre transfer works smoothly, but occasionally there are some anomalies in pronunciation. I'm not sure if it's caused by the dataset of the pre-training model. I would like to try creating a pre-training model based on open source datasets. How can I do this? |
Beta Was this translation helpful? Give feedback.
Answered by
34j
Apr 13, 2023
Replies: 1 comment
-
|
You can just (adjust the hyperparameters and) start training from scratch. When you have finished your training, we would love to have you provide us with your model. (to save money and co2) |
Beta Was this translation helpful? Give feedback.
0 replies
Answer selected by
34j
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
You can just (adjust the hyperparameters and) start training from scratch. When you have finished your training, we would love to have you provide us with your model. (to save money and co2)