KittenTTS ๐ป
KittenTTS is an open-source, state-of-the-art text-to-speech (TTS) model developed by KittenML. It achieves high-quality voice synthesis with only 15 million parameters and a model size of under 25MB, making it one of the most lightweight production-ready TTS systems available.
๐ Key Features
- Ultra-Lightweight: The entire model is under 25MB, smaller than most app icons in enterprise software.
- CPU-Optimized: No GPU required. Runs efficiently on any device, including edge hardware and mobile.
- High-Quality Voices: Supports multiple premium voice options across male and female variants.
- Fast Inference: Optimized for real-time speech generation.
- Simple API: Minimal code required to get started โ install, load model, generate audio.
๐ Quick Start
Installation
pip install https://github.com/KittenML/KittenTTS/releases/download/0.1/kittentts-0.1.0-py3-none-any.whl
Basic Usage
from kittentts import KittenTTS
m = KittenTTS("KittenML/kitten-tts-nano-0.2")
audio = m.generate(
"This high quality TTS model works without a GPU",
voice='expr-voice-2-f'
)
# Save the audio
import soundfile as sf
sf.write('output.wav', audio, 24000)
Available Voices
| Voice ID | Gender |
|---|---|
| expr-voice-2-m | Male |
| expr-voice-2-f | Female |
| expr-voice-3-m | Male |
| expr-voice-3-f | Female |
| expr-voice-4-m | Male |
| expr-voice-4-f | Female |
| expr-voice-5-m | Male |
| expr-voice-5-f | Female |
๐ System Requirements
Works on virtually any platform โ Windows, macOS, Linux โ with no GPU dependency.
๐ Roadmap
- [x] Release a preview model
- [ ] Release fully trained model weights
- [ ] Release mobile SDK
- [ ] Release web version
๐ Project Stats
| Metric | Value |
|---|---|
| Language | Python |
| License | Apache-2.0 |
๐ Links
๐ License
Licensed under the Apache 2.0 License.