gronky_ on AiOla open-sources ultra-fast ‘multi-head’ speech recognition model · Hackerly

Comment on AiOla open-sources ultra-fast ‘multi-head’ speech recognition model

GH repo: https://github.com/aiola-lab/whisper-medusa