This morning Spotify rolled out another way for users to discover new music. It's called Fresh Finds, and it's the official launch of a feature that has actually been on the platform since last year, but wasn't heavily promoted or prominently featured until today.

Fresh Finds is the polar opposite of Discover Weekly, another feature created by an Echo Nest alumni. Discover weekly relies on playlist data from Spotify to generate its suggestions, and each set of tracks is personalized for the individual user. Fresh Finds, by contrast, reaches beyond Spotify to the open web to generate its playlists and isn't personalized at all.

With the launch today, however, Spotify has refined the product to give it broader appeal. Before today Fresh Finds was a mishmash of genres. The algorithms focused on finding hot new tracks, regardless of the style of music, making it most interesting for users with a broad palate not afraid to experiment with new sounds. That original list will continue to exist — it has nearly 100,000 subscribers — but there are also five new categories of Fresh Finds focused on hip hop, electronica, pop, guitar, and experimental ambient sounds.

At this point every major streaming service offers users curated and personalized recommendations. But there is no other service that employs natural language processing across the web to generate its suggested listening. That unique approach to discovery may explain why Spotify, despite its lack of big name exclusives, continues to grow at an increasingly rapid pace, even in the face of an ever more crowded and competitive marketplace.

As streaming services like Apple and Tidal recruit established artists for exclusives on their platform, Brian Whitman, co-founder of The Echo Nest and brains behind Fresh Finds, hopes his algorithms will eventually help Spotify source the best new acts. "Fresh Finds is an amazing tool we use internally for things like artist relations," he told The Verge last September. "We have nothing to announce right now, but I think in the future you’ll see more about how we can use these predictive algorithms to make sure these artists come to Spotify."