Hi!
Could you recommend ways to extract learned audio features in a JS / web environment, client side?
I’ve found:
- and see options for audio classification, but could I skip the classification head, using those APIs, to obtain the feature vector / embedding?