How will Spotify’s mood-spotting feature work?
Did you know that soon you will be targeted with music suggestions based on your mood? We often choose the songs or artists that we listen to based on how we feel or how we want music to make us feel. Soon streaming platform Spotify will offer a helping hand by recommending songs based on your emotions, using a ‘mood-detecting technology’.
How will this new feature work?
Spotify has recently patented a technology that will allow it to analyse your voice as well as any background noise and suggest songs based on your ‘emotional state, gender, age or accent’. It will use voice recognition technology to make observations about your environment and emotional state and offer music suggestions based on this.
The technology would be able to analyse various audio signals picked up by the mic on your smartphone/ tablet/ PC, in order to understand who you are and how you feel. The patent suggests that speech recognition will be used to guess the user’s age and gender by using “a combination of vocal tract length and pitch”, while contextual cues such as ‘intonation, stress and rhythm’ would provide insight as to whether the user is ‘happy, sad, angry or neutral’.
How would Spotify come to these conclusions? The platform will use personal data to crate different personas like extroverts, introverts, etc., so for example, if your tone of voice is more upbeat, high-pitched and exciting you might be assigned a personality trait of extroversion and be offered music based on it.
The results from the voice analysis would be used with other information about the user available on the platform such as previously played songs, most streamed artists, etc. to improve recommendations.
Spotify mentioned that the reason behind considering using speech recognition is to be able to offer more personalised media recommendations to users as the current approach is rather ‘dissatisfactory’.
Like with any other feature that feeds off users’ data, there are of course considerations and concerns around privacy. While Spotify hasn’t yet confirmed whether these features will be implemented in reality, the data privacy aspect would need to be considered ahead of time as well as providing users with the option to opt-out of giving access to their mic and using their data for personalisation.
Here are a few other articles around voice recognition and how your data is handled by apps and voice assistants that you might find interesting: