WEBVTT

00:00:00.326 --> 00:00:01.456 A:middle
&gt;&gt; My name is Ehsan Hoque.

00:00:02.086 --> 00:00:06.286 A:middle
I'm an Assistant Professor of Computer
Science at the University of Rochester.

00:00:07.336 --> 00:00:09.116 A:middle
So I'm fascinated by human emotion.

00:00:09.386 --> 00:00:12.946 A:middle
I want computers to be able to
understand and respond to human emotion,

00:00:13.496 --> 00:00:15.626 A:middle
so there's a basic research going on there.

00:00:16.166 --> 00:00:21.116 A:middle
We publish papers, supervise Ph.D.
students, provide outreach services,

00:00:21.576 --> 00:00:24.766 A:middle
so that's on the research
side, and on the teaching side,

00:00:25.316 --> 00:00:31.146 A:middle
I teach classes on human-computer interaction,
interactive machine learning, exposing new ideas

00:00:31.146 --> 00:00:33.786 A:middle
to the students and inspire their desire.

00:00:34.356 --> 00:00:38.436 A:middle
So I want people to interact
with computers in a natural way.

00:00:39.256 --> 00:00:42.616 A:middle
I want the interaction to be
conversation, just like to be -- to a human.

00:00:43.306 --> 00:00:45.946 A:middle
And how could technology
enable that possibility?

00:00:46.196 --> 00:00:49.736 A:middle
How could technology alleviate
some of the frustration

00:00:49.736 --> 00:00:52.116 A:middle
that we have while we engage
with the technology?

00:00:52.676 --> 00:00:54.676 A:middle
So that one aspect of research.

00:00:55.176 --> 00:00:59.566 A:middle
And the second aspect is: How could this
technology add value to your daily life?

00:01:00.426 --> 00:01:04.866 A:middle
Now, we have built a prototype where now
you can go to a website, turn on your webcam

00:01:05.456 --> 00:01:09.826 A:middle
and practice speaking, and the system
automatically gives you feedback

00:01:09.826 --> 00:01:15.156 A:middle
on your smile intensity, volume
modulation, body language, word usage,

00:01:15.416 --> 00:01:18.096 A:middle
word cloud, and many other dimensions.

00:01:18.506 --> 00:01:22.476 A:middle
Other application could be helping
individuals who may have Asperger's syndrome.

00:01:22.756 --> 00:01:24.706 A:middle
It's a variation of autism.

00:01:25.386 --> 00:01:30.596 A:middle
Many of these individuals love to get some
feedback, but they fear social stigma,

00:01:30.996 --> 00:01:34.456 A:middle
and I think there's a sweet spot for
automated technology to be helpful.

00:01:34.876 --> 00:01:39.226 A:middle
So I teach one class per
semester, so that's twice a week.

00:01:39.946 --> 00:01:44.246 A:middle
I -- before the class, I spend around
three hours preparing for the lecture,

00:01:44.576 --> 00:01:48.586 A:middle
preparing my jokes, making sure
that I have the appropriate breaks.

00:01:49.406 --> 00:01:50.926 A:middle
And teaching is all about engagement.

00:01:50.926 --> 00:01:53.486 A:middle
It's not just I'm there providing information.

00:01:53.936 --> 00:01:57.036 A:middle
It's about: How do I engage the
people that's right in front of me

00:01:57.696 --> 00:02:00.066 A:middle
and then be able to learn together?

00:02:00.316 --> 00:02:03.786 A:middle
And that requires a lot of practice, a lot
of rehearsing, so I do that a lot before --

00:02:03.786 --> 00:02:07.576 A:middle
even though lecture is one hour,
15 minutes, the preparation prior

00:02:07.576 --> 00:02:08.976 A:middle
to that is more than three hours.

