musicube blog image

How can AI help independent artists?

May 12nd, 2022

If you are an artist, you are probably aware of how many aspects of your career are affected by artificial intelligence. For example, you have to deal with discovery algorithms on Spotify, Bandcamp, and other streaming platforms. You know that it's up to Google’s search engine to make your music or website appear on the search results. Finally, you might have experienced YouTube's melody recognition motor. It might have helped you identify if someone else is using your music, or accused you of using someone else's music on your track.

AI seems to be everywhere. And in many cases, it will be acting against you unless you learn how to get it on your side. So this is what this post is about: how to use the AI power on your side.

After reading this article, you will know how to use AI to get more chances to be found in the streaming world. Furthermore, you will understand why AI could be one of the most democratizing tools to break down the barriers between top artists and independents.

Starting from the basics: the world of metadata!

If you haven't heard the word “metadata” before, you are in the right place to get fully into it.

Metadata is the backstage of your digital file: most people never see it, but it is where a significant amount of important stuff happens. So we should take care of it.

Metadata is the information attached to a digital file that does not directly define the music wave but helps identify the file. In its more basic form, a file’s metadata should contain enough information to track you and your music down if there is nothing available other than the digital file: song title, authors, performers, featured artists, release date, and contact information.

But this is not all. When you officially release your track, ideally, an ISRC (International Standard Recording Code) should be generated and attached by the distributor to identify that particular master version. If you are a songwriter collecting money from your compositions, another code, called ISWC (International Standard Musical Work Code), should eventually be attached to your track’s metadata. This code identifies the musical work instead of the recording. Suppose the track ever gets remixed, or a cover version is released. The new track will get a new ISRC code, but it will keep the ISWC. The ISWC will help everybody to know that you are the songwriter of that song, entitled to collect the royalties you are owed from the use of your music.

And we aren't done yet! Have you ever submitted your track for an editorial playlist? The information you fill up, related to music style, lyrics language, main instruments, vocals, mood, and even lyrics, is also attached to your track as metadata.

If some author is missing or misspelled, or a code is typed incorrectly, the revenues will fall into the wrong hands or not be paid at all. Type “The Black Box of the Music Industry” on your browser for more surprising details.

If the information about the music style, language, etc., is not accurate, the track will probably fall on the wrong ears. For example, imagine someone asking for a hard-rock track that is given a Bossanova ballad.

Have you ever seen the librarian's quote: "A misplaced book is a lost book"? Likewise, in digital music files, "An incorrectly tagged track is a lost track."

Next week we will be showing you how the AI could help you to polish your metadata like a pro.