๐ฑ KittenTTS - High Quality Text-to-Speech
Generate high-quality speech from text using KittenTTS, a lightweight TTS model that works without GPU!
Choose from multiple voice options and enter your text to hear the synthesized speech.
๐ค Select Voice
Choose between different male and female voices
๐ Example Texts to Try (Short & Simple):
Click on any example to try it out
| Text to Speech | ๐ค Select Voice |
|---|
About KittenTTS Nano:
- Lightweight 15M parameter text-to-speech model
- Works without GPU - optimized for efficiency
- Multiple voice options (male and female variants)
- 24kHz output sample rate
- Best with short texts (under 400 characters)
- Model: KittenML/kitten-tts-nano-0.1
- Built by KittenML
Usage Tips for Nano Model:
- โ Keep text short and simple (about 400 characters)
- โ Use common words and standard punctuation
- โ Break long content into shorter sentences
- โ Avoid very long sentences or complex punctuation
- โ Avoid technical jargon or unusual words