Our smartphones are already pretty smart, but what if they could detect emotions and suggest appropriate information, activities or media accordingly? Empathetic devices are just round the corner and could have a huge impact on mental health, medical research - as well as suggesting the perfect song to match your mood.
Getting to the heart of human emotion
For a long time now software has been very good at detecting emotions, as seen in this clip of emotion recognition software in action from 2010.
Able to detect real-time micro and multiple facial expressions, this software is actually better at recognising human emotions than humans are – tests showing that it gets it right 65% of the time as opposed to 56% for us mere mortals.
Now imagine how empowering that would be for someone with Asperger’s or autism who finds it extremely difficult to interpret people’s emotions. You could combine that with a heads up display (as with the now discontinued Google Glass or many of the other smart glasses on the market) that tells you to “Tell another one, they liked it!” or “I think they could do with some sympathy.”
However the draw-back with this approach is that you have to point a camera at someone’s face. Whether this is built into smart glasses or you’re actually pointing your phone or wearable camera at them, it’s a pretty intrusive and limiting approach.
Now imagine if emotions could be detected by wireless signals alone.
Smart emotion detection
A new research project from MIT’s Computers Science and Artificial Intelligence lab is able detect emotions with an even greater degree of accuracy than the above software. The device, called the EQ-Radio, uses wireless radio signals to pick up emotions including excitement, sadness, anger and happiness with around 87 percent accuracy.
The system detects subtle cues from the subject including breathing patterns and heart rhythm without the need for any on-body sensors whatsoever.
Emotional intelligence in all your devices
As this approach requires nothing to be worn by the user, and avoids the potential accuracy pitfalls of camera-based facial recognition software (such as the subject being partially or wholly out of view of the camera, being poorly lit, wearing glasses or make-up etc), its creators believe it could be the ideal technology for companies looking to build emotional intelligence into their products.
Professor Dina Katabi, who led the development of the EQ-Radio at MIT, suggests that it could be used in a number of ways across various industries including entertainment, advertising and healthcare sectors.
For example, EQ Radio could be used in smart TVs (or smart set-top boxes such as the Apple TV) to more accurately gauge viewer responses to ads and programming, or built into a smart home hub (such as Google Home) to trigger automated actions with connected devices like stereos and lighting, adjusting the mood of you home to counter or augment your emotions.
A feel-good future
EQ-Radio could eventually find its way into smartphones and even everyday domestic appliances. Having a kettle that can tell jokes when you’re in the mood might not be everyone’s cup of tea, but a self-locking medicine cabinet when you’re feeling low might actually save lives.
My smartwatch can already send my live heartbeat to a loved one and I can foresee a future where we’ll be ever-aware of the emotions of our friends and family and more in-tune with our own.