search

jordipons / musicnn

Pronounced as "musician", musicnn is a set of pre-trained deep convolutional neural networks for music audio tagging.
ISC License
590 stars 87 forks source link

[QUESTION] About models #8

Closed loretoparisi closed 4 years ago

loretoparisi commented 4 years ago

Hello Jordi, I was looking at the recent TensorflowPredictMusiCNN, that's an amazing work, congrats! In the essentia documentation there are several models linked here. Is the same musicnn architecture and models when installing it via pip or it has been updated in the essentia repo?

Thank you!

jordipons commented 4 years ago

Thanks for your interest! I suggest directly asking to Essentia developers (Pablo and Dmitry), who might have the answers to these details!

loretoparisi commented 4 years ago

@jordipons thank you Jordi, I did.