---
license: mit
viewer: false
---
### Introduction
I have created this dataset by taking 50K clean and 50K dirty from people speech and the core purpose is to *fine-tune* seed-vc model.