Spotify patent reveals work on technology to suggest music by monitoring user speech and emotion

In 2018, Spotify filed for a patent that would utilize recordings of users’ speech and background noise to determine specific patterns. From that information, Spotify could then recommend certain music, podcasts, and even advertisements to users.

The patent was approved on January 12, 2021, and it has certainly caught some attention. Taking a closer look, the patent goes over some potential use cases of the unique technology. The most interesting of these is the extraction of intonation, stress, rhythm, and other aspects of a users’ speech to determine their emotional state, gender, age, and accent (via Pitchfork).

Implementing the technology could also allow Spotify to serve better, more relevant advertisements to its users. The patent discusses even being able to discern a users’ environment and if they are alone or with a group of people. All of this voice data would supposedly be retrieved from users’ metadata and then run through a Markov model to make determinations.

Whether or not Spotify decides to move forward with and use the technology described in the patent remains to be seen. Tech companies are famous for frequently filing patents that never actually make it to market. When Pitchfork reached out to Spotify for comment on the patent, this is what they had to say:

“Spotify has filed patent applications for hundreds of inventions, and we regularly file new applications. Some of these patents become part of future products, while others don’t. Our ambition is to create the best audio experience out there, but we don’t have any news to share at this time.”