TTS WebUI

No results found.

Navigation

Home

Pipeline

Text-to-Speech

Text-to-Speech > Kokoro

Text-to-Speech > Chatterbox

Text-to-Speech > Bark

Bark > Generation

Bark > Voice Generation

Bark > Voices

Bark > Settings

Bark > Vocos NPZ

Bark > More Voices

Text-to-Speech > Tortoise

Text-to-Speech > Maha TTS

Text-to-Speech > MMS

Text-to-Speech > Vall-E-X

Audio/Music Generation

Audio/Music Generation > ACE-Step

Audio/Music Generation > MusicGen

Audio/Music Generation > MAGNeT

Audio/Music Generation > Stable Audio

Audio Conversion

Audio Conversion > RVC

Audio Conversion > Demucs

Audio Conversion > Vocos Wav

Outputs

Outputs > History (Slow!)

Outputs > Favorites (Slow!)

Outputs > FFMPEG Metadata

Tools

Tools > GPU Info

Tools > Huggingface Cache Manager

Tools > API Config

GitHub Feedback / Bug reports

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

The pipeline is still a work in progress and might change in future updates.

This pipeline takes an audio file as input and runs it through a generation model to generate a representation of the audio. This representation is then refined by a postprocess model to generate a new audio file.

Choose a generation model:

bark

tortoise

musicgen

magnet

maha

mms

vallex

History Prompt Semantic:

History Prompt:

Voice:

Language:

English

German

Spanish

French

Hindi

Italian

Japanese

Korean

Polish

Portuguese

Russian

Turkish

Chinese

Speaker ID:

0

1

2

3

4

5

6

7

8

9

Use V2

Prompt type:

Short prompt (<15s)

Split prompt by lines

For each subsequent generation:

Use old generation as history

or Use history prompt setting

or Clear history

Text temperature

0.001.20

0.001.20

Waveform temperature

0.001.20

0.001.20

Max length (s)

0.118.0

0.118.0

Seed:Randomize

Burn In Prompt (Optional):

Prompt:

Choose a postprocessing model:

none

rvc

demucs

vocos wav

vocos npz (bark only)