We’ve been having a lot of fun generating music with Suno. It is amazing what their system can do. I first heard about it in a Rolling Stone article ( https://www.rollingstone.com/music/music-features/suno-ai-chatgpt-for-music-1234982307/ ). The article featured a Mississippi blues song about AI generated from a text prompt, “solo acoustic Mississippi Delta blues about a sad AI.” https://soundcloud.com/rs-539916550/soul-of-the-machine . Since then we have made all kinds of songs to test the range of what it can do. I’m sure it will wear off but it is really novel to make custom songs for whatever comes to mind. One of my favorites at the moment is a sea shanty about tax season that my brother made. My son generated a “soothing” metal song. I tried all kinds from blues to big band to metal to bluegrass and was impressed that it was able to make a Gregorian chant with verse in Haiku form. You can even tell it what language to generate a song in. Some of the songs have a story structure and the AI can make appropriate and relevant themes and references to characters in the stories throughout the song. I keep wondering how much of this might be confirmation bias on our part, hearing what we expect to hear and shoehorning parts into an articulated whole, but what this is capable of is impressive. I wonder what directions this might go in, like generating music on the fly based on keywords that come up in conversations. (When I first played with Chat GPT I remember thinking this would be something to use with a search engine because essentially I was using it to find and summarise information; then literally the next day it was announced that it was going to be used with search engines; now it’s time for on-the-fly context-specific background music to make an appearance.) It’s free and available to generate a limited number of songs a day. https://app.suno.ai/
Leave a Reply