Standard for new dataset

[JRMeyer]

>>> namhoai1995
[September 4, 2019, 11:30am]

Hi, I wonder the standard for a new dataset. Like how long, how many
sentences or words does it need?

I want to make a new voice, kids voice and the language still English.
Is it possible to do that by fine-tune the best model with small
dataset?

Thank you and thanks for the hard work.

[This is an archived TTS discussion thread from discourse.mozilla.org/t/standard-for-new-dataset]

[JRMeyer]

>>> erogol
[September 5, 2019, 10:48am]

this might help a bit https://github.com/mozilla/TTS/wiki/Dataset

[Archived Post]