A 164-megabyte speech model transcribes an hour of audio in about twenty seconds on a laptop
Fermion Research built it by compressing Nvidia's 2.5-gigabyte Parakeet model until each encoder weight holds one of five learned levels, at about 2.1 bits.
Company
Fermion Research is an organisation working in artificial intelligence and based in United States. This page collects every Aivio News story about Fermion Research, together with its models and recorded benchmark results.
A 164-megabyte speech model transcribes an hour of audio in about twenty seconds on a laptop
Fermion Research built it by compressing Nvidia's 2.5-gigabyte Parakeet model until each encoder weight holds one of five learned levels, at about 2.1 bits.
Privacy preferences
We use necessary cookies to run this site. With your consent we also use advertising cookies. You can change your choice at any time. Cookie Policy · Privacy Policy