MIT researchers have developed wearable software that reveals a person's emotions based on the way they speak
Software capable of sensing a person's mood. Researchers at Boston's Computer Science and Artificial Intelligence Laboratory have managed to develop an app to be installed in smartwatches that uses algorithms to monitor audio, text transcripts and bio-signals to find out how a person really feels.
The researchers found that certain signals, such as a flat tone of voice, long pauses, increased heart activity, indicate the telling of a sad event. And so far, it didn't take software to guess that. But the software is able to analyze mood even when analyzing just five seconds of conversation - in short, almost in real time - with an accuracy of about 18 percent. The software, at this time, is being tested on a Samsung Simband, but the MIT team aims to integrate it into other commercial devices such as the Apple Watch.
A virtual "social coach"
(Taken from YouTube)
The MIT (Massachusetts Institute of Technology) researchers also plan to take "readings" that reveal many more nuances, capturing, for example, even brief moments when the speaker is tense or excited. Their hope is that someday this type of software could be used in wearable devices as a sort of "social coach" - equipped with artificial intelligence - for people who suffer from anxiety or Asperger's syndrome, that serious disorder characterized by difficulty in social interactions and unusual, narrow patterns of behavior and interests.
Technology for better living
"Imagine if, at the end of a conversation, it was possible to go back and find out at what times the people around us perceived us as the most anxious," explains Tuka Alhanai of the research team. Who goes on to point out that their work "is a step in this direction" and that the time will not be so far away when a wrist-mounted social coach - thanks to the paw of artificial intelligence - can, gradually, suggest us how to deal with the world by keeping our sense of anxiety and our various moods at bay.
Some of the videos in this section were taken from the internet, then rated in the public domain. If the subjects present in these videos or the authors had something against the publication, just make a request for removal by sending an email to: [email protected] We will provide to the cancellation of the video in the shortest time possible.