Sending Audio Features through Machine Learning Model

[john_gorri]

Plan

Free

Country

United States

My Question or Issue

For a personal project, I wanted to use a pre-trained ML model to run predictions off of the audio features of a song. I won't be using the audio features to train the model or even update it in the future, I just want to make predictions based on the audio features of the song. For reference I found an open source dataset on Kaggle that I'll be using to train the model and then using the trained model to create predictions on any new song that I give to the model. Is this against the Spotify developer terms? I want to reiterate that I won't be using and data from the API to train or even update the model.

[csipapicsa]

Hi,

Just a side note: I believe Spotify use these models for extracting audio features: 

https://essentia.upf.edu/models.html