With Spotify having approximately 100 million tracks and more than 600 million subscribers, helping listeners find the music they love has become a navigational challenge. The promise of personalization and meaningful recommendations will give more meaning to the vast catalog and is core to Spotify’s mission.
The streaming audio giant’s suite of recommended tools has evolved over the years: Spotify home feed, discovery weekly, mix, day listand Mix made for you. In recent years, there have been signs that it is working.According to data released by Spotify Investor Day 2022The number of artists Spotify discovers each month has increased from 10 billion in 2018 to 22 billion, and “we’re far from done,” the company said at the time.
Over the past decade or more, Spotify has been investing in artificial intelligence, specifically machine learning. Its recently launched AI DJ may be its biggest bet yet, with the technology enabling subscribers to better personalize listening sessions and discover new music. AI DJs mimic the atmosphere of radio by announcing song titles and track starts, in part to help listeners step out of their comfort zones. One of the existing pain points of AI algorithms—which do a great job of delivering something to an audience that it knows they already like—is predicting when you want to push out of that comfort zone.
AI DJ combines personalization technology, generative AI and dynamic AI sounds, allowing listeners to click the DJ button when they want to hear new music or music that is not directly derived from their established preferences. Behind the sweet tones of AI DJs are tech experts and music experts whose goal is to improve the Spotify tool’s recommendation capabilities. The company has hundreds of music editors and experts around the world. A Spotify spokesperson said generative AI tools allow human experts to “expand their innate knowledge in unprecedented ways.”
Data about a specific song or artist captures a few attributes: a specific musical characteristic, and which song or artist it is typically paired with Artificial intelligence algorithms can access data from millions of listening sessions. Gathering information about a song is a fairly simple process, including year of release, genre, and mood—from upbeat to danceable or melancholy. Various musical attributes such as rhythm, key and instrumentation are also identified. Combining data related to millions of listening sessions with the preferences of other users helps generate new recommendations and enables the leap from aggregated data to individual listener hypotheses.
In its simplest form, “Users who like Y also like Z. We know you like Y, so you probably like Z.” This is how artificial intelligence finds matches. Spotify says it’s working. “Since launching DJ, we’ve found that when DJ listeners hear reviews and personal music recommendations, they More willing to try something new (or listen to a song they might have skipped),” the spokesperson said.
If successful, it won’t just be the audience who will find relief from their pain points. A great discovery tool is equally beneficial for artists looking to connect with new fans.
Julie Knibbe Founder and CEO tomorrow’s music – aims to help artists connect with larger audiences by understanding how algorithms work and how to better use them – says everyone is trying to figure out how to balance familiarity and novelty in a meaningful way, everyone All rely on artificial intelligence algorithms to help make this possible. Be she says the balance between discovering new music and maintaining established patterns is a core yet unresolved issue for everyone involved, from Spotify to listeners and artists.
“Any artificial intelligence is only good at what you tell them to do,” Nibb said. “These recommendation systems have been around for over a decade, and they are very good at predicting what you will like. But they have no way of knowing what is going on in your head, especially when you want to venture into a new musical area or category.”
Spotify day list is an attempt to use generative artificial intelligence to consider established tastes, as well as the different contexts that can shape and reshape listeners’ tastes throughout the day, and come up with new suggestions that fit a variety of moods, activities, and atmospheres. Nibb said similar improvements are likely to continue, with artificial intelligence getting better at finding formulas for how much novelty listeners want, but she added, “The assumption that people always want to discover new music is just not true.”
Most people are still quite happy to return to familiar musical territory and listening patterns.
“It’s different audiences, curators, experts… people have different requirements for artificial intelligence,” Nibb said. “Experts are harder to surprise, but they’re not the majority of listeners, they tend to be more casual,” she said, and their use of Spotify often amounts to creating a “comfortable background” for everyday life.
Technological optimists often speak of an era of abundance. While there are 100 million songs to choose from, many listeners prefer the same 100 songs a million times over, so it’s easy to see why a new balance is being sought. But Ben Ratliff, a music critic and author of “Every Song Ever Written: Twenty Ways to Listen in an Age of Musical Abundance,” says algorithms aren’t so much about solving the problem as Further solidifying this issue.
“Spotify is great at capturing the emotions of the masses and creating a soundtrack for them,” Ratliff said. “It’s Sadgirl Starter Pack For example, the playlist has a great name and about a million likes. Unfortunately, under the banner of The Gift, SSP reduces the oceanic complexity of youthful depression to a handful of reliably “craving” musical performances and allows hard clichés of music and emotion to develop much faster.
Ratliff favors curatorial works that are clearly made by real people with actual preferences. Even a good playlist might be made without much intention or conscience, but just a developed awareness of pattern recognition, “whether it’s an obscure pattern or a well-known pattern,” he said. explain.
Depending on the individual, artificial intelligence may have an equal chance of being a utopian or dystopian solution in a 100 million orbital universe. Ratliff said most users should keep things simpler in their streaming music journey. “As long as you realize that apps will never know you the way you want them to, and as long as you know what you’re looking for, or have some good tips ready, you can find a lot of great music on Spotify.”
About The Author
