Phonon-2 Packs Speech Recognition into 164MB

x.com57 saved

  • post
  • Saved

Written by AI

TLDR
Phonon-2 is a 164MB open-weight speech model claimed to beat Whisper large on average and transcribe an hour in 20 seconds on MacBook Air.

postThe source

Manan Gupta@yoitsmanan

Introducing Phonon-2: a new standard in speech recognition per byte.

At just a tiny 164MB download, it is more accurate on average than OpenAI's Whisper large ( a model 10X its size)

Transcribe an hour of audio in just 20 seconds on a MacBook Air!

Open weights, CC-BY-4.0,

Picture attached to the post by Manan Gupta

Media saved