Spotify is one of the most popular music streaming services in the world. The company is always looking for ways to improve the listener experience and get subscribers to listen to more music. Spotify has announced that it has patented technology allowing it to analyze the voice and suggest songs based on the listener’s emotional state, gender, age, or accent.
Spotify originally filed for the patent in 2018 and was granted on January 12. The patent outlines a method of making observations about the user environment and emotions using speech recognition technology. The idea is that the technology would analyze the user and then play music reflecting the mood or the social setting.
According to the patent, the tech would be able to determine if the user was alone, in a small group, or at a party. Spotify sees the new technology as a significant improvement in the current process of recommending titles that requires users to answer multiple questions about age, gender, and their favorite brands.
The patent indicates Spotify would use contextual clues like intonation, stress, and rhythm to determine if the user was happy, angry, sad, or neutral. The streaming company is clear that the example metadata categories including emotions, gender, age, and accent are examples, and there are numerous other characterizations and classifications it can use.
Data would be combined with other information like the user’s previously played songs and their friend’s taste in music. Spotify has been busy patenting new technology to help improve service. Back in September, it patented a karaoke-like feature allowing people to overlay a music track with their own vocals. It was also recently awarded a patent for Cadence-based media content selection matching the music’s tempo to running speed.