Kitten TTS is an open-source realistic text-to-speech model with just 15 million parameters, designed for lightweight deployment and high-quality voice synthesis.
Currently in developer preview
For custom support - fill this form
Email the creators with any questions : [email protected]
- Ultra-lightweight: Model size less than 25MB
- CPU-optimized: Runs without GPU on any device
- High-quality voices: Several premium voice options available
- Fast inference: Optimized for real-time speech synthesis
pip install https://github.com/KittenML/KittenTTS/releases/download/0.1/kittentts-0.1.0-py3-none-any.whl
from kittentts import KittenTTS
m = KittenTTS("KittenML/kitten-tts-nano-0.1")
audio = m.generate("This high quality TTS model works without a GPU", voice='expr-voice-2-f' )
# available_voices : [ 'expr-voice-2-m', 'expr-voice-2-f', 'expr-voice-3-m', 'expr-voice-3-f', 'expr-voice-4-m', 'expr-voice-4-f', 'expr-voice-5-m', 'expr-voice-5-f' ]
# Save the audio
import soundfile as sf
sf.write('output.wav', audio, 24000)
Works literally everywhere
- Release a preview model
- Release the fully trained model weights
- Release mobile SDK
- Release web version