Beyond the buzzwords
Responsible Dataset Sourcing
Every voice used in our models is ethically licensed and securely sourced by Kits via the artists themselves. No unauthorized scraping of artists' work.
$
Artists directly benefit from the use of their vocal data in our artist model library through dynamic revenue-sharing models.
Our tools give vocalists and producers more control over their sound while opening up new possibilities for revenue streams and creativity.
Fairly Trained AI Models
Kits' voice and instrument models accessible via the model select modal, have been Fairly Trained certified, but using these models with the text to speech functionality is not covered by Fairly Trained. Fairly Trained certification is granted to models that aren't trained on copyrighted work without consent - you can read more about the certification here. Our certification reflects our commitment to training data practices that are fair to creators.
Title: CSTR VCTK Corpus: English Multi-speaker Corpus for CSTR Voice Cloning Toolkit
Author: Christophe Veaux and Junichi Yamagishi and Kirsten MacDonald
Year: 2017
Source: Access here