Lilypad-kokoro-module

This repo represents the hexgrad/kokoro model as a module to be run on the Lilypad Network to produce text-to-audio inference

Inputs

The module allows you to pass in a piece of text and generate an audio sample using the following parameters

• PROMPT -> The text to generate the audio sample from
• VOiCE -> The voice to be used to read the text with the avaailable voices being: heart, fenrir, bella and puck

How to run

Run this on the lilypad network via:

lilypad run gihub.com/Lilypad-Tech/lilypad-kokoro-module:<add-commit-hash-here> -i prompt="hello world" -i voice="heart"

or through the Anura API

Notes

• This model does not support creating custom voices or creating a copy of your own voice

Acknowledgments

• hexgrad/kokoro model