Hugging Face details Phonon on-device text-to-speech launch
Reported by Hugging Face1 source ↓
Hugging Face says Phonon runs without a network on phones, computers and browsers. The company reports more than 200,000 downloads in its first two weeks and says the models ranked near the top of its Open TTS Leaderboard tests.
Key points
- Phonon comes in two models, about 65 MB and 75 MB, and supports five languages
- Phonon 31m has an Apache 2.0 license; Phonon 40m is licensed for non-commercial use
- Hugging Face reports more than 200,000 downloads in the first two weeks
- Hugging Face says Phonon ranked first in German and Spanish intelligibility and in the top three across all evaluated languages
Details
Phonon generates speech locally, without sending text to a server, and works offline, according to Hugging Face. The company says it runs on phones, laptops, single-board computers and in browsers. The models are about 65 MB and 75 MB, and together support English, French, German, Spanish and Portuguese, with more than 100 built-in voices.
Hugging Face says Phonon 31m is available under the Apache 2.0 license, including for commercial use. Phonon 40m is available for non-commercial projects; the company says commercial pricing for the other models is based on the number of users rather than speech volume. Developers can also create custom voices through Gradium Studio.
Hugging Face reports that Phonon was downloaded more than 200,000 times in its first two weeks. On the company's Open TTS Leaderboard, it says the models ranked first for intelligibility in German and Spanish and in the top three in every language evaluated. Hugging Face also says they led its CPU streaming test for time-to-first audio.
Written from coverage as of
Images
- Hugging Face
- Hugging Face