KittenML has released KittenTTS 2 Mini, a text-to-speech model with 82 million parameters, supporting 16 languages and 118 voices. The model is available in various formats, including ONNX with int8 or int4 weights for CPU-only execution, and float32 PyTorch weights for GPU usage. It can be integrated into projects using the KittenTTS library, with usage examples provided for generating speech in different languages and voices. AI
IMPACT Provides a compact, multi-lingual TTS solution for developers, enabling broader integration of speech synthesis capabilities.
RANK_REASON Release of a new, smaller TTS model with multiple language and voice options. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Hugging Face Trending Models →
- Apache Software License 2.0
- Google Colab
- Kaggle
- KittenML/kitten-tts-2-mini
- KittenTTS
- ONNX
- ONNX Runtime
- PyTorch
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →