2 open source projects by BrightGu

Any-to-any voice conversion using synthetic specific-speaker speeches as intermedium features

Any-to-one voice conversion using the data augment strategy: pitch shifted and duration remained.