Generative artificial intelligence is currently one of the hottest topics in Silicon Valley, and its impact is already being felt in the music industry. BandLab — the music-creation app that has become popular on TikTok — relies on AI as the engine for its tool SongStarter. Users can lean on it to generate beats or melodies at random, or prompt it to spit something out based on specific lyrics and emojis; BandLab’s 60 million registered creators are churning out more than 17 million songs each month, including breakout hits for dv4d and ThxSoMch.
The tracks that emerge from BandLab depend on the interaction of human creators and AI. That holds true for some of the companies focusing on functional audio as well. LifeScore, which uses AI to “create unique, real-time soundtracks for every journey,” relies on “Lego blocks of sound all made in a studio by real musicians playing real instruments through lovely microphones,” says co-founder/CEO Philip Sheppard. Even the sound of a stream trickling through a forest comes from “someone going out with a rig and standing in that stream and recording it.”
The AI kicks in when it comes to assembling that sonic Lego. “The AI is saying, ‘Hey, wouldn’t it be delightful if these could arrange themselves in this different way?’” Sheppard explains. “’How about if we could turn that into eight hours that felt like it was original every time you listened to it?’”
All results of these processes may not work. “Unsuccessful soundscapes are generated all the time,” says Oleg Stavitsky, co-founder/CEO of Endel, which offers an app that generates music designed to help users focus, relax or sleep. “Each soundscape goes through a multi-step testing process: from automated testing, detecting sound artifacts and bad sound combinations to in-house testing to our community testing.” That community includes some 4,000 people who provide feedback through Endel’s Discord channel.
“We put human eyes on everything before it goes out,” says Alex Mitchell, founder/CEO of Boomy, a company that offers aspiring musicians the chance to make songs in seconds with help from AI tools. Since 2019, Boomy users have created over 12 million songs. “We have a generic content policy that basically means if all you’re doing is pressing buttons and we detect that, then your release probably won’t be eligible for distribution,” says Mitchell. “We reject way more releases than what gets submitted. That way we’re not flooding the [digital service providers] with a bunch of nonsense.”
How will Boomy scale this approach as it attracts even more users and generates even more millions of songs? “We’re hiring,” Mitchell says.
