Originally Posted By: shlind
Nowadays with all the AI support software popping up everywere, maybe PG Music could link into some simple "AI Package" that can analyze the “Feel” of song

+1 on this.
If we could drag-in an MP3 file of a song whose style we want to replicate, and have BiaB choose a style and/or create a style that matches the MP3, that would be great smile

I'm thinking this would be a big challenge from a computer science/programming perspective and a challenge from a musicology perspective. But I think it is possible. One would have to numerically define what a "style" is, most likely in several dimensions, then determine the style when presented with a specific audio source and then hunt for a similar or exact style in the database. If the style of interest does not exist in the database (which will happen sooner or later) then BiaB would create it so the user can use it.

Edit: I performed a DuckDuckGo search and it yielded a paper that seems to address this very question.

The paper aims to discuss a method for autonomously analyzing a musical style. The algorithm used needs to be shown both positive and negative examples of the style one is trying to teach it. It learns to distinguish the desired musical style from melodies that belong to it and melodies that do not belong to it. The information obtained can later be used for recognizing and/or generating melodies that fit within the analyzed style, a capability that can be of much value in musical creativity.

Warning: This paper is not intended for non-technocrats but the PG Music Development Team may find this to be a useful launching pad if they aren't already aware of it.

https://www.designsociety.org/download-publication/32529/music_style_analysis_using_the_random_forest_algorithm


Last edited by Bass Thumper; 08/05/22 08:29 AM.
https://soundcloud.com/user-646279677
BiaB 2024 Windows
For me there’s no better place in the band than to have one leg in the harmony world and the other in the percussive. Thank you Paul Tutmarc and Leo Fender.