This repository aims at providing efficient CNNs for Audio Tagging. We provide AudioSet pre-trained models ready for downstream training and extraction of audio embeddings.
Why do you think that https://github.com/sigsep/open-unmix-pytorch is a good alternative to EfficientAT