tts_gallery / README.md
Michael Hu
feat: add Kokoro-82M TTS model support
8829e6c
---
title: TTS Galary
emoji: 📣
colorFrom: purple
colorTo: pink
sdk: gradio
sdk_version: 5.44.1
app_file: app.py
pinned: true
---
# TTS Galary
This demo showcases the multilingual capabilities of multiple TTS models, supporting both English and Chinese languages.
## Features
- Text-to-speech generation for English and Chinese
- Gradio web interface for easy interaction
- Real-time audio generation and playback
- Example texts for quick testing
- Support for multiple TTS architectures including seq2seq models
## Requirements
- Python 3.8 or higher
- Required Python packages (automatically installed by Hugging Face):
- chatterbox-tts
- gradio
- torchaudio
- torch
## Usage
1. Enter text in the input box
2. Select the language (English or Chinese)
3. Click "Generate Speech"
4. Listen to the generated audio
## Supported Languages
- English
- Chinese
## Supported Models
- **Chatterbox**: Industrial-grade multilingual TTS solution
- **KittenTTS**: High-quality TTS with voice cloning capabilities
- **Piper**: Local on-device TTS with multiple voice options
- **Faster Whisper**: High-performance speech recognition model for audio transcription
- **Kokoro**: Lightweight TTS model with 82M parameters, Apache-licensed for production and personal use
## Examples
The interface includes example texts for both languages to help you get started quickly.
## Notes
- The first generation may take a moment as the model loads
- Subsequent generations will be faster
- For best results, use clear and properly punctuated text