Hugging Face Blog·· 15 h ago
A Closer Look at Phonon TTS. Launching High-Quality On-Device Speech That Tops the Benchmarks.
A Closer Look at Phonon TTS. Launching High-Quality On-Device Speech That Tops the Benchmarks.
AI summary
Two weeks ago, we opened up Phonon, our small text-to-speech model that runs on your own device. It needs no server and no network, and what you type stays with you.
Selection record
Threshold 60Official, first-handFirst 58Second 71
AdmittedSum of both 129 ≥ twice the threshold 120
- Source tier
- Official, first-hand; this tier's threshold is 60
- Pre-filter
- passed：端侧TTS模型发布与评测
- Why it was chosen
- The official retrospective on the on-device TTS model Phonon gives its size, languages, licence and leaderboard results, showing whether a small model can replace cloud speech.
A model scores each item twice, independently, against one written standard, out of 100. An item is admitted only when the two scores add up to twice the threshold. The threshold is set per source tier.
Source: Hugging Face Blog · huggingface.co