MENU

When user interfaces dig into your emotions

When user interfaces dig into your emotions

Technology News |
By eeNews Europe



Successfully crowd-funded through an IndieGOGO campaign in March last year, the EmoSPARK Cube is described as the world’s first emotionally intelligent home console dedicated to your happiness, no less.

Designed to play a central role in the home and to serve as a voice-controlled universal remote for all other Bluetooth or WiFi-connected electronic devices, the 90x90x90mm cube could not only help users access the information they want in a timely manner, but it could also tune into the owner’s mood to offer comfort with upbeat conversation or even playful suggestions.

The secret recipe to this new venture resides in the patent-pending Emotion Processing Unit (EPU), in effect a microcontroller programmed to create a synthesised emotional response in AI-enabled machines, based on user input such as facial expressions (through their smartphone or tablet camera), or speech tonality and voice inflexions.

The EPU algorithms enable machines to respond to stimuli in line with one of the eight primary emotions: anger, fear, sadness, disgust, surprise, anticipation, trust, and joy, as identified by Robert Plutchik’s psycho-evolutionary theory.

As the EmoSPARK Cube interacts with her/his owner, it stores the perceived emotional stimuli within its memory bank, as emotional patterns or fingerprints. This, claims EmoSHAPE, is possible thanks to the EPU’s Computational Emotional Neural Frequency Architecture (CENFA) and its Emotional Profile Graph (EPG) computation functionality.

Over time, the so-called EPG is used to register and develop a bank of emotional associations for each memory (data) within each Cube. In principle, this emotional profiling could help the Cube share adequate data with other AI technologies, so the user would experience adequate emotional responses from other AI interfaces.

But why a physical cube when pretty much all of this could be done in software in today’s smartphones or at least on the cloud?

“Because we need to establish trust, each cube develops its own emotional intelligence in relation with each specific user, the data connections that are made, the Emotional Profile Graph are very personal”, told us Fitzpatrick.

“The EmoSPARK Cube learns a lot about you, it may even figure out emotional states and data about you that you may not actively know and that you may not want to share with everyone. Hence each cube acts like a mini-server where the EPG is securely hosted” he added.

According to Fitzpatrick, having a standalone product makes the interface more tangible, something that users would want to give a name and engage with in small talk.

“The cube can find information from different sources of knowledge such as Wikipedia, Freebase and internet search engine Google, but also social networks, and it is capable of two-way conversation in tune with the owner’s mood, so it is much more than just a multimedia servant”, continued Fitzpatrick.

EmoSPARK has a conversational engine of more than two million lines of data, the company claims, and every time you chat with EmoSPARK it will learn to develop its own conversational understanding based on past interactions. Even if you were trying to have an argument with the box, it would not get upset but would talk you into a smoother mood.


But the whole idea really is to be able to turn-on and off things more easily around the home, or at least more efficiently than with a remote, with upbeat suggestions lined up or playing emotional tricks to lift the user’s mood.

In effect, guided by your inputs and past history, the machine will adapt its suggestions and conversation until it can improve your mood and read through a happy face.

Now for the “be happy” part, would the EmoSPARK Cube encourage you to have a glass of wine or to frantically eat chocolate cake or even take drugs, if from your history, these were perceived as your best mood enhancers?

Could users sue the company for being encouraged to please themselves beyond socially accepted norms?

“This is a very tricky task, off course we wouldn’t want to encourage illegal or unhealthy behaviours”, commented Fitzpatrick, short of a clear answer. Adding emotional intelligence to fitness tracking devices could definitely encourage users to build a fitness plan and take action to feel better about themselves, but should an electronic device (be it emotionally intelligent) discourage anyone from boredom or even idle time?

Just for the sake of collecting “happy points” on your face, wouldn’t this sort of in-house multimedia assistant push you towards more multimedia content consumption or gaming at the expense of unconnected activities such as outdoor sports or off-screen social events?

The Cube’s insatiable quest for your happiness won’t stop there. Once a reliable EPG is established, it can also “talk” to other Cubes about its experiences through a specially-designed grid via EmoSHAPE’s servers, enabling Cubes with similar affinities to connect and share media bearing similar emotion tags.

More pragmatically, for Fitzpatrick, there is already a number of inanimate objects that people feel close to, such as cars or smartphones, adding emotional artificial intelligence to the interface could make these objects more like friends that can help you do things more naturally.

“In the end, it’s all about having a better user experience, but of course it could go much further than that” said Fitzpatrick.

In the future, the company could consider licensing its EPU algorithms to smart TV manufacturers or to telecom providers willing to bundle the EmoSPARK Cube with their Wi-Fi hub. Surely they would want to figure out the viewers’ emotional response to the content being served, for adaptive scheduling and content creation (adverts, TV series etc…).

Would this figure out that watching TV news is just a horrible experience most of the time?

Visit EmoSHAPE at www.emoshape.com

Related articles:

How will deep learning change SoCs?

Websites’ emotional impact revealed by webcams

Depth cameras interpret shoppers’ behaviour

Lost in Big Data: digital zombies

If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :    eeNews on Google News

Share:

Linked Articles
10s