SYLIN

Phonon

by Fermion Research · Model · Voice & audio

Phonon is Gradium’s text-to-speech family. It turns written text into spoken audio as it generates, making it useful for reading assistant replies,…

AccessOpen weightsNeeds~0.1 GB (the Ollama build)LicenseCC-BY-4.0

Evidence

  1. Oct 4
    Open speech-to-text model under 900 MB that transcribes an hour of audio in about 20 seconds
  2. Oct 4
    Second-lowest average error rate in its benchmarks despite being far smaller than rivals; free Hugging Face spaces to try

Cost to run