Spotify Will Now Listen To Your Voice To Know Your Emotions

Written by | February 3, 2021 0:26 am | No Comments

Spotify wants to know your emotions

Spotify wants to know your emotions


The world is changing very fast, and A.I. is becoming more and more present in everybody’s life. This is obviously scary and has many implications, mostly related to the heavy surveillance of our daily lives. According to Musicbusinessworldwide, Spotify has secured a patent to monitor the background noise and speech of its users. If you are using Spotify, you know that the streaming platform already makes suggestions according to your musical tastes, and probably knows a lot about you, already… but they want to know even more.

The company wants to use speech recognition to determine users’ ‘emotional state, gender, age, or accent.’ ‘Identification of taste attributes from an audio signal’ is a new patent describes as a ‘method for processing a provided audio signal that includes speech content and background noise’ and ‘identifying playable content, based on the processed audio signal content.’

Usually, we provide plenty of information to companies, information such as gender, age, but this goes further and refines the information to better identify the user’s tastes. ‘Intonation, stress, rhythm and the likes of units of speech,’ as well as ‘acoustic information within a hidden Markov model architecture’ allow Spotify ‘to categorize a user’s mood as ‘happy, angry, sad or neutral.’ Yes, Spotify wants to know your emotions.

Spotify will collect data such as the user’s voice and background noise — ‘sounds from vehicles on a street, other people talking, birds chirping, printers printing — to attribute an emotional state using the complex Parrott’s emotional framework. Then ‘prosodic information (e.g., intonation, stress, rhythm and the like of units of speech) can be combined and integrated with acoustic information within a hidden Markov model architecture,’ which is a complex statistical model. This technology can even approach your age and gender, based on a ‘combination of vocal tract length and pitch.’

Once all this information is extracted, Spotify will recommend specific content to the listener, while taking into consideration the listener’s ‘previous requests,’ ‘listening and rating history’ and ‘links to associated profiles such as those of the user’s friends or colleagues,’ as well as ‘the user’s existing music collection and/or library.’ Isn’t it ‘beautiful’? I am joking…

Since when listening to music has been under such high surveillance? Since when using a streaming platform has become signing with Big Brother? Isn’t it terrifying to think Spotify will analyze your voice and listen to your home’s noises? It’s an insidious way to invade people’s privacy but I don’t see why people would not participate in this huge data collection by the most powerful streaming platform… Looking at the success of Amazon’s Alexa, another very invasive technology that plenty of people have accepted without even thinking twice, there is no reason to think that this new Spotify feature will be a problem.

I use Spotify sporadically, I certainly don’t have a paid account and never listen to their recommendations. Why? Because I don’t care about their robotic algorithm. Why would music fit my mood and environment? I actually dislike this idea, it reminds me about Fran Lebowitz lamenting about people who complain they don’t recognize themselves in a book: ‘A book is not supposed to be a mirror. It’s supposed to be a door,’ she said. That’s the attitude I myself recommend instead of listening to the same music over and over. Plus, what else can Spotify do with this impressive amount of data they will collect? What else can they do with all these features on voice and emotion recognition? We certainly have the right to worry about that.


Leave a Reply

Your email address will not be published. Required fields are marked *