TTS WebUI

GitHub Feedback / Bug reports

Home Pipeline Text-to-Speech Audio/Music Generation Audio Conversion Outputs Tools

Kokoro Chatterbox Bark Tortoise Maha TTS MMS Vall-E-X

Generation Voice Generation Voices Settings Vocos NPZ More Voices

VRAM	Advice
≥ 10GB	Use large models
< 10GB	Use small models
< 4GB	Use CPU offloading (requires restart)
< 2GB	Use CPU offloading and only small models

Text generation and coarse-to-fine are of similar importance.
Small models might not have a perceptible difference.

Changes will be automatically saved