Inference rework and noise reduction in preprocess - Githubissues

Vidalnt / Applio

A simple, high-quality voice conversion tool focused on ease of use and performance.

https://applio.org

MIT License

1.58k stars 254 forks source link

Inference rework and noise reduction in preprocess #690

Closed ShiromiyaG closed 1 week ago

ShiromiyaG commented 1 week ago

Description

This PR reworks much of the inference code

Motivation and Context

How has this been tested?

Screenshots (if appropriate):

Types of changes

[ ] Bug fix (non-breaking change which fixes an issue)
[X] New feature (non-breaking change which adds functionality)
[ ] Breaking change (fix or feature that would cause existing functionality to not work as expected)

Checklist:

[X] My code follows the code style of this project.
[ ] My change requires a change to the documentation.
[ ] I have updated the documentation accordingly.
[ ] I have added tests to cover my changes.
[X] All new and existing tests passed.

AznamirWoW commented 1 week ago

Updated inference code to do all manipulations in memory without saving temporary files, split audio no longer recursively calls voice conversion with repeated pre-post processing, batch audio conversion simply calls regular voice conversion in a loop instead of duplicating the code.