Noised data for Finetune

[yadinbenyamin]

Hi there,

I got a lot of telephone emotion recordings and we want to finetune the model in order to add more voices and emotions. Is there any good way that you can think of that will help in order to get some good results (other the just cleaning the noise)

#119

When using the webpage-based demo by running streamlit run demo_page.py, the generated audio contains no noise. However, I do notice noise at the beginning of the sample audio. Can you please provide more details about this issue?

Thanks :)