Spotify secures a patent that allows it to suggest music based on mood
As per the patent, filed in February 2018 and granted in January this year, Spotify wants to get inside a user’s head, analysing their current emotional state, gender, age or even accents.
January 29, 2021 / 04:38 PM IST
Spotify India | Representative Image.
Spotify, the music streaming app, has recently been granted a patent for technology that enables it to analyse voice data. The data to be analysed is collected via speech recognition which will theoretically allow Spotify to detect a user’s mood, which in turn will allow the app to suggest music tailored to the user's mood.
As discussed on a recent post by Music Business Worldwide, the variables collected by the new system will be used to personalise Spotify’s recommendations. These recommendations could range from music and podcast content to ads.
As per the patent, Spotify wants to get inside a user’s head, analysing their current emotional state, gender, age or even accents. The patent was filed in February 2018 and was granted in January this year.
The patent states that Spotify may have a “method for processing a provided audio signal that includes speech content and background noise” that will help in “identifying playable content, based on the processed audio signal content”.
Spotify says the technology enables it to obtain “intonation, stress, rhythm and the likes of units of speech and acoustic information within a hidden Markov model architecture". This then allows the app to categorise the user’s mood as "happy, angry, sad or neutral”.
Spotify will also analyse ambient noise, deciphering “sounds from vehicles on a street, other people talking, birds chirping, printers printing, and so on”.
This is not the first patent filing that has cropped up in recent times. In September, Spotify filed a patent for a new feature that allows listeners to overlay their own vocals over a music track. Another filing in November was for geo-targeted advertising that will be tailored for the listener using data collected with the help of 3D audio.