A group of researchers have developed a new MIT wearable app that is artificially intelligent. It can determine the emotions in every conversation as it can predict tone from the speech, pattern and vitals of a person.
As reported in zdnet, it is considered as a potential future coach for people with Aspergers or issues in picking up social cues and emotions. This MIT wearable app was made using a wearable device that can pick high-resolution waveforms in terms of physiological metrics including blood pressure and heart rate. The device that is used is the Samsung Simband.
People with the mentioned anxieties usually have a hard-time handling social interactions. But with the MIT wearable app, it may help to make every interaction better with the system's goal to make everything easier to understand and measure.
According to Tuka Al Hanai, a graduate student and a co-author of a related paper, "Imagine if, at the end of a conversation, you could rewind it and see the moments when the people around you felt the most anxious," She also added that the work that they've made is a step in a direction suggesting that they may not be that far from a world of people that can have an AI social coach with the MIT wearable app.
Scientists trained two algorithms after collecting not less than 30 conversations. This step is to classify all the nature of conversations whether it's happy, sad, positive, negative or neutral. With this small MIT wearable app, a lot of things from emotions can now be explained.
Based on the report from mashable, the machine has 83% of accuracy that can indicate monotonous vocal tones, long pauses, and a twitch that has something to do with sad stories, while energetic, varied patterns of speech to identify happy stories. With this MIT wearable app, a lot of people suffering from anxiety can be provided with its assistance. But since it is not yet ready, people might have to wait a little bit more longer.