Skip to Content
77 Mass Ave

Robo-Coach

Software helps hone social skills.

New software developed at MIT can help people who have difficulty with interpersonal skills practice until they feel more comfortable with situations such as a job interview or a first date. The software, called MACH (for “my automated conversation coach”), uses a computer-generated onscreen face, along with facial, speech, and behavior analysis and speech synthesis software, to simulate face-to-face conversations. It then gives users feedback on their interactions.

Robo-Coach
MACH, automated conversation coaching software, lets users practice social skills in their own environment and at their own pace. They can practice as much as they wish—and own the resulting data.

The research was led by MIT Media Lab doctoral student M. Ehsan Hoque, PhD ’13, who says the work could be helpful to a wide range of people, including those with Asperger’s syndrome. “Interpersonal skills are the key to being successful at work and at home,” Hoque says. With his system, he says, people have full control over the pace of the interaction and can practice as many times as they wish. And in randomized tests with 90 MIT juniors, the software showed its value.

It runs on an ordinary laptop, using its webcam and microphone to analyze the user’s smiles, head gestures, and speech volume and speed, among other things. The automated interviewer can smile and nod in response to the subject’s speech and motions, ask questions, and give responses.

Each participant had two simulated job interviews, a week apart, with different MIT career counselors. Between those interviews, the students were split into three groups. One watched videos of general interview advice; the second had a practice session with MACH and watched a video of their performance but got no feedback; a third group used MACH and then saw videos of themselves along with graphs showing when they smiled, how well they maintained eye contact, how well they modulated their voices, and how often they used filler words such as “like,” “basically,” and “umm.”

Evaluations showed statistically significant improvement by members of the third group on measures such as whether the counselor would “recommend hiring this person.” There was no significant change for the other two groups.

One reason the automated system gives effective feedback, Hoque believes, is precisely that it’s not human: “It’s easier to tell the brutal truth through the system because it’s objective.”

At press time Hoque expected to earned a doctorate in media arts and sciences at MIT over the summer and become an assistant professor of computer science at the University of Rochester this fall.

Keep Reading

Most Popular

A Roomba recorded a woman on the toilet. How did screenshots end up on Facebook?

Robot vacuum companies say your images are safe, but a sprawling global supply chain for data from our devices creates risk.

A startup says it’s begun releasing particles into the atmosphere, in an effort to tweak the climate

Make Sunsets is already attempting to earn revenue for geoengineering, a move likely to provoke widespread criticism.

10 Breakthrough Technologies 2023

Every year, we pick the 10 technologies that matter the most right now. We look for advances that will have a big impact on our lives and break down why they matter.

These exclusive satellite images show that Saudi Arabia’s sci-fi megacity is well underway

Weirdly, any recent work on The Line doesn’t show up on Google Maps. But we got the images anyway.

Stay connected

Illustration by Rose Wong

Get the latest updates from
MIT Technology Review

Discover special offers, top stories, upcoming events, and more.

Thank you for submitting your email!

Explore more newsletters

It looks like something went wrong.

We’re having trouble saving your preferences. Try refreshing this page and updating them one more time. If you continue to get this message, reach out to us at customer-service@technologyreview.com with a list of newsletters you’d like to receive.