search

NVIDIA / flowtron

Flowtron is an auto-regressive flow-based generative network for text to speech synthesis with control over speech variation and style transfer
https://nv-adlr.github.io/Flowtron
Apache License 2.0
889 stars 177 forks source link

Changing speaker Voice #99

Closed shehrum closed 3 years ago

shehrum commented 3 years ago

Hi team,

Is it possible to change the voice of the TTS model i.e. adapt/finetune it to a new dataset with a different voice such as in this paper: Cascading ASR + TTS. From what I read in the documentation and samples, its possible to do style transfer, but didn't find anything on voice conversion.

rafaelvalle commented 3 years ago

yes, look at the Fine-tuning for few-shot speech synthesis in the README.