What are the audio features to best describe a music?

Every cepstral coefficients can be considered as one of the best features for defining a musical piece.

Most famous being the Mel Scale, as I can see you are already extracting MFCCs, you are good to go. Although you should have mentioned, which MFCC are you extracting, from experience (a little bit) first 15 are usually the most useful cause they have a positive value. You can also work on more intrinsically robust GFCC.

They can then be used to obtain Spectrograms and so on.

I would suggest not using kNN, prefer Random Forrest, moreover audio signals require a lot of preprocessing. DCT and STFT are a must.

The features you have selected are a good starting point, but are still (with the exception of tempo) quite "low level" compared to what might be most relevant for music recommendation systems.

The Essentia project provides feature extractors for music, that cover both low-level, medium-level and (since Jan 2020) high-level music feature descriptors. Their high-level descriptors include:

• musical genre
• ballroom music classification
• moods: happy, sad, aggressive, relaxed, acoustic, electronic, party
• western / non-western music
• tonal / atonal
• danceability
• voice / instrumental
• gender (male, female singer)
• timbre (dark, bright)

The medium and low-level descriptors cover all what you mention, and more. This is packaged into a command-line tool that outputs JSON.