Excerpt of article by Ann Robinson | The Guardian
Ellie is on the screen, she’s welcoming and gentle with a soft encouraging voice. She asks probing questions in a non-judgmental way. Her body language is encouraging and her face is interested and responsive. She’s obviously highly trained and skilful at picking up cues.
But Ellie isn’t real. She’s a virtual interviewer created by a team of scientists at the Institute for Creative Technologies (ICT) at the University of Southern California. She may be an avatar, but she’s great to talk to.
Ellie introduces herself and asks a series of questions that can pick up signs of depression and post-traumatic stress disorder, or PTSD. A sensor and webcam scan your facial and body movements and tone of voice, and instruct Ellie how best to interact with you. She’s not listening just to the words, but also judging the tone and facial expression that accompany the words.
Watch the following video to see how the technology performs.
To read the whole article by Ann Robinson, click here.
Thanks to the Institute for Creative Technologies (ICT) at the University of Southern California for creating Ellie; Ann Robinson for writing the article; The Guardian for committing its resources to publishing the article; Google for helping me find the article; YouTube for hosting the video; and all the people who, directly or indirectly, made it possible for me to include the picture, video, and text in this post.