FermionResearch / Phonon-2
Audio10.2K380Updated Oct 23, 2026
Downloads
10.2K
Likes
380
Last updated
Oct 23, 2026
License
cc-by-4.0
Phonon-2 is the most accurate open speech recognition model for English under 900 MB. Across the Open ASR Leaderboard's seven English sets it averages 5.21 % word error, beating models multiple times its size in raw bytes. Set for set it holds the accuracy of the 2.5 GB full-p…
Tasks & library
- automatic speech recognition
- mlx
Tags
- mlx
- parakeet_tdt_five_value
- apple-silicon
- speech-to-text
- asr
- stt