Phonon

Open speech recognition models for English, running on a laptop or datacenter GPU.

Servicefreeglobal

Phonon is a set of free, open speech recognition models for English from Fermion Research, small enough to run on a laptop or a datacenter GPU with a CLI and CPU/CUDA images, transcribing an hour of audio in about two and a half minutes. It's aimed at developers who want local, open speech-to-text, similar in role to Whisper, instead of relying on a cloud speech API.

Website

Full match profile

Behind the summary, Matchbox keeps a richer profile of Phonon - the signals our matcher actually reads to decide when to surface it. It stays private; claim the listing to see and control it.

Problem & pain-point mapping
Who we surface it to (audience fit)
What it's a strong alternative to
Trust & credibility signals

Claim Phonon to unlock →

Something wrong with this listing — dead link, not a real product, wrong info?

Try Matchbox with your own problem

Describe what is not working - we’ll show you whether Phonon (or something else) actually fits.

Try a problem