Spotify granted patent that could monitor users’ speech to make music recommendations
Spotify has secured a patent which would allow it to use technology in order to monitor users’ speech and collect data so as to customise music recommendations.
As Music Business Worldwide reports, the patent was approved earlier this month on January 12, after the streaming giant filed for it back in 2018.
In its application, which you can read here, Spotify say the technology would work by retrieving audio, including voice signals and background noise, to understand “content metadata” about users such as their emotional state, gender, age and accents.
The patent goes on to say that the platform would rely on information such as “intonation, stress, rhythm and the likes of units of speech” to determine if a user is feeling “happy, angry, sad or neutral”.
It would also retrieve information about users’ environments. The filing refers to both physical environments like on public transport or outdoors, as well as social environment – that is, whether the user is alone, in a small group, or at a party.
Environmental metadata that would be gathered could include “sounds from vehicles on a street, other people talking, birds chirping, printers printing, and so on”.
In the filing’s description section, Spotify says that in “the field of on-demand media streaming services, it is common for a media streaming application to include features that provide personalized media recommendations to a user.”
It goes on to say that one current approach to identifying a user’s taste is through asking questions about basic information such as their gender and age, along with additional information such as the kinds of artists a user likes, which further “fine-tune[s] the system’s identification of their taste”.
Spotify says that this system, as it requires users to “tediously input answers to multiple queries” in order for the algorithm to understand their taste.
“What is needed is an entirely different approach to collecting taste attributes of a user, particularly one that is rooted in technology so that the above-described human activity (e.g., requiring a user to provide input) is at least partially eliminated and performed more efficiently.”
In a statement given to Pitchfork earlier today, a representative for the company commented: “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.”
In November, Spotify announced it was testing a new “Discovery Mode” feature which would allow artists to boost visibility of specific tracks through the platform’s recommendation algorithms, in exchange for a “promotional recording royalty rate” on those songs’ streams.
The news arrived shortly after the United Musicians and Allied Workers Union launched a ‘Justice at Spotify’ campaign demanding, among other things, that all artists receive at least one cent per stream of their songs on the platform.