Gradium says Phonon, its open text-to-speech models, can generate speech directly on phones, laptops and other devices without a network connection. The models come in two sizes, about 65 MB and 75 MB, and together support English, French, German, Spanish and Portuguese, with more than 100 built-in voices.
On Hugging Face’s Open TTS Leaderboard, Gradium says Phonon ranked first for intelligibility in German and Spanish, placed in the top three in every language it evaluated, and led the CPU streaming test for time-to-first audio. The company says Phonon 31m is Apache 2.0 licensed for commercial use; Phonon 40m is limited to non-commercial use.
Why it matters
App developers can use Phonon 31m to add offline speech without sending users’ text to a server, though the company’s leaderboard results are not independently validated here.
