microsoft / CLAP

Learning audio concepts from natural language supervision
MIT License
455 stars 35 forks source link

difference between version 2022 and 2023 #32

Closed sivannavis closed 4 months ago

sivannavis commented 4 months ago

Hi! Just wanna confirm, according to the two papers, is the 2022 version HTSAT-22+GPT2 and the 2023 version CNN14+BERT? Thank you!

sivannavis commented 4 months ago

Oh I saw the configs, seem like so.