The gradium/phonon-31m-en model is a high-quality, on-device text-to-speech system capable of generating natural 24 kHz English speech. It is small enough to run in a browser, on a phone, or a laptop CPU, and streams audio during generation. This model builds upon Pocket TTS and the Mimi codec, developed by Kyutai, and is available under an Apache 2.0 license. AI
IMPACT Enables high-quality, on-device speech generation for applications requiring local processing.
RANK_REASON The item describes a specific model release with instructions for use, fitting the 'tool' category.
Read on Hugging Face Trending Models →
- AI Engineer Podcast
- Latent Space Podcast
- Easton
- GitHub
- Google Colab
- gradium/phonon-31m-en
- Kaggle
- Kyutai
- Mimi codec
- Phonon
- Pocket TTS
- transformers
AI-generated summary · Google Gemini · from 3 sources. How we write summaries →