TTS WebUI

GitHub   Feedback / Bug reports
HomePipelineText-to-SpeechAudio/Music GenerationAudio ConversionOutputsTools
KokoroChatterboxBarkTortoiseMaha TTSMMSVall-E-X

Text-to-Speech Models:

Kokoro

Kokoro is a fast and lightweight TTS model with 82 million parameters. Small but comparable in quality to larger models.

Run Github

Chatterbox

Expressive text-to-speech model with reference audio support for voice cloning.

Run Github

Bark

Bark is a text-to-speech model that can generate speech from text.

Run Github

Tortoise

Tortoise is a text-to-speech model that can generate speech from text.

Run Github

Maha TTS

Maha TTS is a text-to-speech model that can generate speech from text, supports many Indian languages.

Run Github

MMS

Fairseq based text-to-speech model that supports 1000+ languages

Run Github

VALL-E X

Multilingual TTS: Speak in three languages - English, Chinese, and Japanese - with natural and expressive speech synthesis.

Run Github