search

Gadersd / whisper-burn

A Rust implementation of OpenAI's Whisper model using the burn framework
MIT License
263 stars 29 forks source link

Tinygrad dependancy for weights conversion #16

Open DavidGOrtega opened 11 months ago

DavidGOrtega commented 11 months ago

May I ask why the need of Tinygrad for the weights conversions? The script seems to be dumping them with np afterwards is read by tinygrad.

Gadersd commented 11 months ago

I used the tinygrad whisper implementation as a reference when porting since it's concise. There's no reason the script couldn't be easily modified to only use torch for loading the weights.