With about 100 million tracks available and over 600 million subscribers, helping listeners find the music they are going to love has turn out to be a navigational challenge for Spotify. It is the promise of personalization and meaningful recommendations that may give the vast catalog more meaning, and that’s central to Spotify’s mission.
The streaming audio giant’s suite of suggestion tools has grown through the years: Spotify Home feed, Discover Weekly, Mix, Daylist, and Made for You Mixes. And lately, there have been signs that it’s working. According to data released by Spotify at its 2022 Investor Day, artist discoveries every month on Spotify had reached 22 billion, up from 10 billion in 2018, “and we’re nowhere near done,” the corporate stated at the moment.
Over the past decade or more, Spotify has been investing in AI and, particularly, in machine learning. Its recently launched AI DJ could also be its biggest bet yet that technology will allow subscribers to higher personalize listening sessions and discover recent music. The AI DJ mimics the vibe of radio by announcing the names of songs and lead-in to tracks, something aimed partially to help ease listeners into extending out of their comfort zones. An existing pain point for AI algorithms — which could be excellent at giving listeners what it knows they already like — is anticipating whenever you want to break out of that comfort zone.
The AI DJ combines personalization technology, generative AI, and a dynamic AI voice, and listeners can tap the DJ button once they want to hear something recent, and something less-directly-derived from their established likes. Behind the dulcet tones of an AI DJ there are people, tech experts and music experts, who aim to improve the suggestion capability of Spotify’s tools. The corporate has lots of of music editors and experts across the globe. A Spotify spokesperson said the generative AI tool allows the human experts to “scale their innate knowledge in ways never before possible.”
The info on a selected song or artist captures a couple of attributes: particular musical features, and which song or artist it has been typically paired with among the many hundreds of thousands of listening sessions whose data the AI algorithm can access. Gathering information in regards to the song is a reasonably easy process, including release 12 months, genre, and mood — from completely satisfied to danceable or melancholic. Various musical attributes, reminiscent of tempo, key, and instrumentation, are also identified. Combining this data related to hundreds of thousands of listening sessions and other users’ preferences helps to generate recent recommendations, and makes the leap possible from aggregated data to individual listener assumptions.
In its simplest formulation, “Users who liked Y also liked Z. We know you want Y, so you may like Z,” is how an AI finds matches. And Spotify says it’s working. “Since launching DJ, we have found that when DJ listeners hear commentary alongside personal music recommendations, they’re more willing to try something recent (or listen to a song they might have otherwise skipped),” the spokesperson said.
If successful, it is not just listeners that get relief from a pain point. An ideal discovery tool is as useful to the artists looking for to construct connections with recent fans.
Julie Knibbe, founder & CEO of Music Tomorrow — which goals to help artists connect with more listeners by understanding how algorithms work and the way to higher work with them — says everyone seems to be trying to determine how to balance familiarity and novelty in a meaningful way, and everyone seems to be leaning on AI algorithms to help make this possible. Be she says the balance between discovering recent music and staying with established patterns is a central unresolved issue for all involved, from Spotify to listeners and the artists.
“Any AI is barely good at what you tell them to do,” Knibbe said. “These recommender systems have been around for over a decade and so they’ve turn out to be superb at predicting what you’ll like. What they can not do is know what’s in your head, specifically whenever you want to enterprise out right into a recent musical terrain or category.”
Spotify’s Daylist is an attempt to use generative AI to keep in mind established tastes, but in addition the various contexts that may shape and reshape a listeners’ tastes across the course of a day, and make recent recommendations that fit various moods, activities and vibes. Knibbe says it’s possible that improvements like these proceed, and the AI gets higher at finding the formula for a way much novelty a listener wants, but she added, “the idea that individuals want to discover recent music on a regular basis shouldn’t be true.”
Most individuals still return, fairly happily, to familiar musical terrain and listening patterns.
“You’ve gotten various profiles of listeners, curators, experts … people put different demands on the AI,” Knibbe said. “Experts are tougher to surprise, but they don’t seem to be nearly all of listeners, who tend to be more casual,” and whose Spotify usage, she says, often amounts to making a “comfortable background” to every day life.
Technology optimists often speak when it comes to an era of “abundance.” With 100 million songs available, but many listeners preferring the identical 100 songs 1,000,000 times, it is easy to understand why a recent balance is being sought. But Ben Ratliff, a music critic and writer of “Every Song Ever: Twenty Ways to Listen in an Age of Musical Plenty,” says algorithms are less solution to this problem than an extra entrenching of it.
“Spotify is nice at catching onto popular sensibilities and making a soundtrack for them,” Ratliff said. “Its Sadgirl Starter Pack playlist, as an illustration, has an incredible name and about 1,000,000 and a half likes. Unfortunately, under the banner of a present, the SSP simplifies the oceanic complexity of young-adult depression right into a small collection of dependably ‘yearny’ music acts, and makes hard clichés of music and sensibility form more quickly.”
Works of curation which can be clearly made by actual individuals with actual preferences remain Ratliff’s preference. Even a great playlist, he says, might need been made without much intention and conscience, but only a developed sense of pattern recognition, “whether it’s patterns of obscurity or patterns of the broadly known,” he said.
Depending on the person, AI can have equal probabilities of becoming either a utopian or dystopian solution inside the 100-million track universe. Ratliff says most users should keep it more easy of their streaming music journeys. “So long as you realize that the app won’t ever know you in the way in which you would like to be known, and so long as you know what you are in search of, or have some good prompts on the ready, you could find a number of great music on Spotify.”