README.md

metadata

title: Kokoro TTS
emoji: ⚡
colorFrom: purple
colorTo: purple
sdk: gradio
sdk_version: 6.5.1
app_file: app.py
python_version: 3.11
pinned: false

Kokoro TTS & Voice Cloning System

A complete solution for high-quality Text-to-Speech using the Kokoro-82M model.

Features

• All Pre-built Voices: Includes American, British, and International voices (Spanish, French, Italian, Japanese, Portuguese).
• Long Script Processing: Automatic chunking and merging for processing long texts without cutting off.
• Voice Cloning Interface: Upload a reference to "clone" voices (supports loading custom .pt voice vectors).
• Gradio Web UI: Simple and intuitive interface for all functions.

Installation

1. Run setup.bat to install all dependencies.
2. Make sure you have Python installed and it's in your PATH.

How to Run

1. Run run.bat.
2. Open the URL provided in the terminal (e.g., http://127.0.0.1:7860).

Voices Included

• American: af_heart, af_bella, af_nicole, af_sarah, af_sky, am_adam, am_michael, am_fenrir, am_puck.
• British: bf_emma, bf_isabella, bm_george, bm_lewis.
• International: jo_zeus (JA), es_barcelona (ES), fr_paris (FR), it_florence (IT), and more.

Long Scripts

Toggle "Enable Long Script Processing" for large texts. This will split the text into chunks, process them individually, and merge the result into a single continuous audio file.

Check out the configuration reference at https://huggingface.co/docs/hub/spaces-config-reference