↓

Research Hugging Face ·

A Closer Look at Phonon TTS: High-Quality On-Device Speech That Tops the Benchmarks

Hugging Face blog post on Phonon TTS, described in its title as high-quality on-device speech that tops the benchmarks.

What the article says

Gradium's Phonon is a small open-weight text-to-speech model that runs entirely on-device, with no server or network. Two sizes (about 65 and 75 MB) cover English, French, German, Spanish and Portuguese, with 100+ built-in voices plus voice cloning and design in Gradium Studio. Phonon 31m (English) is Apache 2.0; the others are CC BY-NC. It was downloaded over 200,000 times in two weeks, ranks first in German and Spanish intelligibility on Hugging Face's Open TTS Leaderboard and tops the CPU streaming benchmark for time to first audio (28 ms). It runs on phones, laptops, browsers and NPUs via the ptts framework, and is in Transformers from v5.21.

Read the official article AnnuIA profile

Summary written by AnnuIA in its own words from the official page of Hugging Face. The content and rights belong to the publisher: please read the original article.

More announcements from Hugging Face

Latest AI news

All news