Giving a voice to the unheard | Rabea Ziuod | TEDxJerusalem
-
0:08 - 0:09Hi.
-
0:11 - 0:14We are living in an exciting era,
-
0:14 - 0:19where innovation and technology
has the potential to do the unimaginable, -
0:19 - 0:23and it becomes even more unimaginable
-
0:23 - 0:27when it breaks down the gaps
between disability and ability. -
0:28 - 0:3115% of the world population
-
0:33 - 0:35- 1 billion people around the world -
-
0:35 - 0:37lives with disabilities
-
0:37 - 0:42which makes people with disabilities
the largest minority in the world. -
0:43 - 0:45And they are not living
on a different planet. -
0:45 - 0:50They may be part of our families,
friends, or colleagues. -
0:51 - 0:56Today, I'm going to tell you
how people with speech disabilities -
0:56 - 0:59will have a way to better communicate.
-
0:59 - 1:03I was 7 years old
when my sister Amal was born. -
1:03 - 1:06I was too young to see the challenges
-
1:06 - 1:09that my family was facing
on a daily basis, -
1:09 - 1:14but I could see that Amal
couldn't crawl, or eat, or talk -
1:14 - 1:17like any other baby her age.
-
1:17 - 1:22But with time, we adjusted
to raise a baby with cerebral palsy, -
1:22 - 1:26while understanding her special
communication patterns and needs. -
1:28 - 1:30Nine years later,
-
1:30 - 1:33my family was blessed
to have another baby, Ahmad. -
1:34 - 1:38Ahmad decided to grow up
exactly like his sister Amal, -
1:38 - 1:43being so smart, so sharp,
curious about everything around him, -
1:43 - 1:47but he also decided to invent
his special communication patterns -
1:47 - 1:49to communicate with us,
-
1:50 - 1:53and for the other people
that couldn't understand him, -
1:53 - 1:55we had to translate.
-
1:55 - 2:00Amal and Ahmad say "num"
when they are hungry, -
2:00 - 2:05and they say "ahh" to call
the name of Nora, my sister. -
2:05 - 2:09And when they want to call
my name, they say "abeya". -
2:09 - 2:13In case they want to go
to the bathroom, they say "kkhh". -
2:13 - 2:17We understand most
of their special communication patterns, -
2:17 - 2:21but it's only us, the close circle.
-
2:21 - 2:25And this is the case for most
of the people who have an unclear voice. -
2:26 - 2:29One of those people is Urit.
-
2:29 - 2:34Urit is a 34-year-old woman
with cerebral palsy. -
2:34 - 2:36She is living an independent life.
-
2:36 - 2:41She can drive her car, go to the gym,
and do a lot of other things. -
2:43 - 2:48However, when it comes
to communicating using her voice, -
2:48 - 2:51sometimes, it can become
harder than going to the gym, -
2:51 - 2:53and more frustrating
-
2:53 - 2:59because she finds herself repeating
the same words again and again -
2:59 - 3:01in order to be understood.
-
3:01 - 3:05We asked Urit to say
a few words in English. -
3:06 - 3:08Let's listen to her together
-
3:08 - 3:11and see if you can understand
what she's trying to say. -
3:12 - 3:14(unclear speech)
-
3:17 - 3:22I don't know how many of you
could understand her this first time, -
3:22 - 3:23but let's listen to her again,
-
3:23 - 3:28and really focus and try to understand
what she's trying to say. -
3:28 - 3:29(unclear speech)
-
3:33 - 3:37Try to memorize what she has just said;
we'll get to that later. -
3:39 - 3:42With my siblings, and Urit,
and people that I get to know, -
3:42 - 3:46I had the chance to see
a world full of challenges, -
3:46 - 3:49- a world of special people with needs.
-
3:50 - 3:54And this allowed me
to examine the existent technology -
3:54 - 3:58in search of an answer
for what my siblings were seeking. -
3:59 - 4:02Unfortunately, the current
state of the art assistive technology, -
4:02 - 4:07including speech recognition applications,
could not provide an answer. -
4:08 - 4:14Since then, all the assistive technology
has completely bypassed the voice, -
4:14 - 4:17opting to use other modes of communication
-
4:18 - 4:22[by] replacing the voice
with symbols and images, -
4:22 - 4:26or movements of the body
in the head or in the eyes. -
4:27 - 4:32This brings me to the other lightweight
alternative that does use the voice -
4:33 - 4:36which is speech recognition applications.
-
4:36 - 4:39This technology works in two approaches.
-
4:40 - 4:44The first approach attempts
to discover which word has been said. -
4:46 - 4:49The second approach relies on phonemes.
-
4:49 - 4:54Phonemes are all the sounds
we produce using our mouth and nose. -
4:56 - 5:00Both approaches rely on statistical models
-
5:00 - 5:03from a large database of standard speech.
-
5:03 - 5:06But once the speech is not standard,
-
5:06 - 5:10- when I say not standard,
I mean it's enough to have an accent, -
5:10 - 5:12like most of us here -
-
5:12 - 5:14this will not work.
-
5:14 - 5:20My colleagues and I developed
a new approach of assistive technology -
5:20 - 5:22that does use the person's own voice
-
5:22 - 5:26and can understand
non-standard speech patterns, -
5:26 - 5:32with the mission to give people
with a speech disability their voice back. -
5:33 - 5:36So, whose life is this going to change?
-
5:36 - 5:39People with cerebral palsy,
-
5:39 - 5:42people with Parkinson's,
and Myasthenia Gravis, -
5:42 - 5:44so many [other] neurological disorders,
-
5:44 - 5:47people who are born
with hearing disabilities, -
5:47 - 5:52or people who suddenly have a stroke
and their whole life is changed, -
5:52 - 5:55but not only theirs.
-
5:55 - 5:59Not only the people who have
difficulty expressing themselves, -
5:59 - 6:03but everyone who interacts
with them on a daily basis. -
6:03 - 6:09This will make it easier
for them to be socially included -
6:09 - 6:13- because every one of us
wants to be socially included. -
6:13 - 6:18And now, you may be asking yourself,
"How does it work?" -
6:18 - 6:22"How come the current speech recognition
technology couldn't do the same?" -
6:25 - 6:28Because our technology works
in a different way. -
6:29 - 6:32So, each person has to go
through two phases. -
6:32 - 6:35The first phase is called
the calibration phase, -
6:35 - 6:41where the person has to teach the device
and the application his own patterns -
6:41 - 6:44by entering the patterns
and building his own dictionary. -
6:44 - 6:46This phase usually happens
-
6:46 - 6:49with the person
who understands him the most. -
6:49 - 6:51Together they will build the dictionary.
-
6:51 - 6:55This generally takes
only one to three hours, -
6:55 - 6:58and it depends on the speaking
capability of the speaker. -
6:58 - 7:00After building the dictionary,
-
7:00 - 7:04we move to the second phase
which is the recognition phase. -
7:04 - 7:08The application will be able to recognize
unintelligible speech patterns -
7:08 - 7:11from the dictionary that is already built
-
7:11 - 7:14and translate them
into a clear voice in real time. -
7:16 - 7:20Our approach is user-dependent
and language-independent -
7:20 - 7:23which means it can work
in any language in the world, -
7:24 - 7:26even the invented ones.
-
7:26 - 7:30And the key word here
is 'pattern-matching'. -
7:30 - 7:35Once the person builds his own dictionary,
and says a word that already exists there, -
7:35 - 7:37there will be a pattern-matching
-
7:37 - 7:40between what he says
and what it already exists. -
7:40 - 7:42But here we found a problem.
-
7:42 - 7:45We found that people
with a speech disability -
7:45 - 7:48pronounce different words in similar ways.
-
7:50 - 7:54And the challenge was
to differentiate between them. -
7:54 - 7:57So we created a technology
called Adaptive Framing. -
7:58 - 8:04Adaptive Framing technology can be adapted
to the width of the event in the pattern. -
8:04 - 8:10In the existing technology, you can see
the L and the A in the same frame. -
8:10 - 8:15But in our new technology, you can see
that the L and A are in different frames -
8:15 - 8:18which increases the accuracy
of the pattern-matching. -
8:19 - 8:22And this makes
our pattern-matching so much better. -
8:23 - 8:26I suppose you still remember Urit, right?
-
8:26 - 8:31Let's listen to her again now,
but this time using Talkitt: -
8:34 - 8:35(unclear speech)
-
8:35 - 8:36Now I can ...
-
8:36 - 8:37(unclear speech)
-
8:37 - 8:38... start
-
8:38 - 8:40(unclear speech)
-
8:40 - 8:41... speaking freely.
-
8:43 - 8:45(Applause)
-
8:56 - 8:58Talkitt is only one step
-
8:58 - 9:02towards bridging the gap
between disability and ability -
9:02 - 9:05by letting people express their potential.
-
9:05 - 9:07The more we challenge our minds,
-
9:07 - 9:12the more gaps will collapse
to let us all have a normal life. -
9:12 - 9:13Thank you.
-
9:13 - 9:14(Applause)
- Title:
- Giving a voice to the unheard | Rabea Ziuod | TEDxJerusalem
- Description:
-
This talk was given at a local TEDx event, produced independently of the TED Conferences.
Talkitt is developing an innovative solution that will enable people that suffer from motor, speech, and language disorders to easily communicate using their own voice, by translating the unintelligible pronunciation into understandable speech.
- Video Language:
- English
- Team:
- closed TED
- Project:
- TEDxTalks
- Duration:
- 09:21
Denise RQ approved English subtitles for Giving a voice to the unheard | Rabea Ziuod | TEDxJerusalem | ||
Denise RQ edited English subtitles for Giving a voice to the unheard | Rabea Ziuod | TEDxJerusalem | ||
Denise RQ edited English subtitles for Giving a voice to the unheard | Rabea Ziuod | TEDxJerusalem | ||
Denise RQ edited English subtitles for Giving a voice to the unheard | Rabea Ziuod | TEDxJerusalem | ||
Denise RQ edited English subtitles for Giving a voice to the unheard | Rabea Ziuod | TEDxJerusalem | ||
Denise RQ edited English subtitles for Giving a voice to the unheard | Rabea Ziuod | TEDxJerusalem | ||
Denise RQ edited English subtitles for Giving a voice to the unheard | Rabea Ziuod | TEDxJerusalem | ||
Denise RQ edited English subtitles for Giving a voice to the unheard | Rabea Ziuod | TEDxJerusalem |