CLAP is to audio what CLIP is to image. This is an improved CLAP checkpoint, specifically trained on music.
Browse all Feature Extraction models →