Rana el Kaliouby: This app knows how you feel — from the look on your face

Prevodilac: Marija Kojić Lektor: Tijana Mihajlović

Our emotions influence every aspect of our lives,

Naše emocije utiču na svaki aspekt naših života,

from our health and how we learn, to how we do business and make decisions,

od našeg zdravlja i načina učenja

do načina obavljanja posla i donošenja odluka,

big ones and small.

bile one velike ili male.

Our emotions also influence how we connect with one another.

Naše emocije takođe utiču na to kako se povezujemo jedni sa drugima.

We've evolved to live in a world like this,

Razvili smo se za život u svetu kao što je ovaj,

but instead, we're living more and more of our lives like this --

ali umesto toga, živimo sve više naše živote ovako -

this is the text message from my daughter last night --

ovo je SMS poruka koju mi je ćerka poslala sinoć -

in a world that's devoid of emotion.

u svetu koji je lišen emocija.

So I'm on a mission to change that.

Tako sam na misiji to da promenim.

I want to bring emotions back into our digital experiences.

Želim da vratim emocije u naše digitalno iskustvo.

I started on this path 15 years ago.

Krenula sam ovim putem pre 15 godina.

Bila sam kompjuterski naučnik u Egiptu

I was a computer scientist in Egypt,

and I had just gotten accepted to a Ph.D. program at Cambridge University.

i samo što su me primili

na doktorski program na univerzitetu u Kembridžu.

So I did something quite unusual

Tako sam uradila nešto prilično neobično

for a young newlywed Muslim Egyptian wife:

za jednu mladu, tek venčanu ženu koja je muslimanska Egipćanka:

(Smeh)

With the support of my husband, who had to stay in Egypt,

uz podršku mog muža koji je morao da ostane u Egiptu,

I packed my bags and I moved to England.

spakovala sam kofere i preselila se u Englesku.

At Cambridge, thousands of miles away from home,

Na Kembridžu, kilometrima daleko od kuće,

I realized I was spending more hours with my laptop

shvatila sam da više vremena provodim sa laptopom

nego sa bilo kojim drugim ljudskim bićem.

than I did with any other human.

Yet despite this intimacy, my laptop had absolutely no idea how I was feeling.

Ipak, i pored ove prisnosti, moj laptop nije imao pojma kako se osećam.

It had no idea if I was happy,

Nije imao pojma da li sam srećna,

having a bad day, or stressed, confused,

da li sam imala loš dan, da li sam pod stresom, zbunjena,

and so that got frustrating.

tako da je to postalo frustrirajuće.

Even worse, as I communicated online with my family back home,

Još gore, dok sam komunicirala sa porodicom kod kuće preko mreže,

I felt that all my emotions disappeared in cyberspace.

osećala sam da sve moje emocije nestaju u sajber prostoru.

I was homesick, I was lonely, and on some days I was actually crying,

Nedostajala mi je kuća, bila sam usamljena i nekih dana sam zapravo i plakala,

but all I had to communicate these emotions was this.

ali sve što sam imala da iskažem ove emocije bilo je ovo.

(Laughter)

(Smeh)

Today's technology has lots of I.Q., but no E.Q.;

Današnja tehnologija ima mnogo IQ-a, ali nema EQ;

mnogo kognitivne inteligencije, ali nimalo emocionalne inteligencije,

lots of cognitive intelligence, but no emotional intelligence.

So that got me thinking,

te me je to nagnalo na razmišljanje.

what if our technology could sense our emotions?

Šta ako bi naša tehnologija mogla da oseti naše emocije?

What if our devices could sense how we felt and reacted accordingly,

Šta ako bi uređaji mogli da prepoznaju i reaguju na naša osećanja,

just the way an emotionally intelligent friend would?

baš kao što bi reagovao i emocionalno inteligentan prijatelj?

Those questions led me and my team

Ta pitanja su navela mene i moj tim da napravimo tehnologiju

to create technologies that can read and respond to our emotions,

koja može da čita i reaguje na naše emocije,

and our starting point was the human face.

a naša početna tačka bilo je ljudsko lice.

So our human face happens to be one of the most powerful channels

Dakle, ljudsko lice je jedno od najmoćnijih kanala

that we all use to communicate social and emotional states,

koje koristimo da prenesemo društvena i emocionalna stanja,

everything from enjoyment, surprise,

sve od zadovoljstva, iznenađenja,

empathy and curiosity.

empatije do radoznalosti.

In emotion science, we call each facial muscle movement an action unit.

U nauci o emocijama,

sve pokrete facijalnih mišića nazivamo akcijskim jedinicama.

So for example, action unit 12,

Tako, na primer, akcijska jedinica 12 nije holivudski blokbaster,

it's not a Hollywood blockbuster,

it is actually a lip corner pull, which is the main component of a smile.

nego se radi o podizanju ugla usana, što je glavna komponenta osmeha.

Try it everybody. Let's get some smiles going on.

Probajte to. Hajde da svi nabacimo osmehe.

Another example is action unit 4. It's the brow furrow.

Još jedan primer je akcijska jedinica 4. To je boranje obrve.

It's when you draw your eyebrows together

To je kada skupite obrve i stvorite sve ove teksture i bore.

and you create all these textures and wrinkles.

Ne volimo ih, ali one su jasan pokazatelj neke negativne emocije.

We don't like them, but it's a strong indicator of a negative emotion.

Tako imamo oko 45 ovih akcijskih jedinica

So we have about 45 of these action units,

and they combine to express hundreds of emotions.

i one se kombinuju da izraze stotine emocija.

Teaching a computer to read these facial emotions is hard,

Podučavanje kompjutera da čita ove facijalne emocije je teško

because these action units, they can be fast, they're subtle,

zato što ove akcijske jedinice mogu biti brze, suptilne

and they combine in many different ways.

i kombinuju se na mnogo različitih načina.

So take, for example, the smile and the smirk.

Uzmite tako, na primer, osmeh i zloban osmeh.

They look somewhat similar, but they mean very different things.

Izgledaju pomalo slično ali imaju veoma različita značenja.

(Laughter)

(Smeh)

So the smile is positive,

Dakle, osmeh je pozitivan,

a smirk is often negative.

a zloban osmeh je često negativan.

Ponekad jedan zloban osmeh može da vas napravi poznatim.

Sometimes a smirk can make you become famous.

But seriously, it's important for a computer to be able

Ozbiljno, važno je da kompjuter bude u stanju

to tell the difference between the two expressions.

da prepozna razliku između ova dva izraza.

So how do we do that?

Dakle, kako to postižemo?

We give our algorithms

Obezbeđujemo svojim algoritmima

tens of thousands of examples of people we know to be smiling,

desetine hiljada primera ljudi za koje znamo da se osmehuju,

from different ethnicities, ages, genders,

ljudi različitih etničkih pripadnosti, godina, različitog pola,

and we do the same for smirks.

a to isto činimo za podsmehe.

And then, using deep learning,

Onda, uz dubinski pristup učenju,

the algorithm looks for all these textures and wrinkles

algoritam traži sve ove teksture i bore

and shape changes on our face,

i promene oblika na našim licima,

and basically learns that all smiles have common characteristics,

i u suštini uči da svi osmesi imaju zajedničke osobine,

all smirks have subtly different characteristics.

da se svi zlobni osmesi suptilno razlikuju od osmeha po osobinama.

And the next time it sees a new face,

I sledeći put kada vidi novo lice,

it essentially learns that

u suštini uči da ovo lice ima iste osobine osmeha

this face has the same characteristics of a smile,

and it says, "Aha, I recognize this. This is a smile expression."

i kaže „Aha, prepoznajem ovo. Ovo je izraz osmeha.”

So the best way to demonstrate how this technology works

Najbolji način da pokažemo kako ova tehnologija funkcioniše

is to try a live demo,

jeste da probamo demo-verziju uživo, tako da mi treba dobrovoljac.

so I need a volunteer, preferably somebody with a face.

Poželjno je da taj neko ima lice.

(Smeh)

(Laughter)

Cloe's going to be our volunteer today.

Kloi će nam danas biti dobrovoljac.

So over the past five years, we've moved from being a research project at MIT

Dakle, u zadnjih pet godina, od istraživačkog projekta na MIT-u

to a company,

postali smo kompanija,

where my team has worked really hard to make this technology work,

u kojoj je moj tim vredno radio da ova tehnologija uspe,

kako mi volimo da kažemo, u divljini.

as we like to say, in the wild.

And we've also shrunk it so that the core emotion engine

Takođe smo je smanjili tako da osnovni emotivni motor

radi na bilo kom mobilnom uređaju koji ima kameru, kao što je ovaj Ajped.

works on any mobile device with a camera, like this iPad.

So let's give this a try.

Dakle, hajde da probamo.

As you can see, the algorithm has essentially found Cloe's face,

Kao što možete da vidite, algoritam je pronašao Kloino lice.

so it's this white bounding box,

To je ovaj beli granični okvir,

and it's tracking the main feature points on her face,

koji prati glavne tačke odlika na njenom licu,

so her eyebrows, her eyes, her mouth and her nose.

dakle, njene obrve, njene oči, njena usta i njen nos.

The question is, can it recognize her expression?

Pitanje je, da li može prepoznati njen izraz lica?

So we're going to test the machine.

Dakle, testiraćemo mašinu.

Pre svega, da vidimo tvoje pokeraško lice. Da, super.

So first of all, give me your poker face. Yep, awesome. (Laughter)

(Smeh)

And then as she smiles, this is a genuine smile, it's great.

Onda, kada se osmehne, ovo je iskren osmeh, odlično.

So you can see the green bar go up as she smiles.

Vidite da se zelena traka puni kada se osmehuje.

Now that was a big smile.

To je bio širok osmeh.

Can you try a subtle smile to see if the computer can recognize?

A jedan suptilan osmeh da vidimo da li kompjuter ume da prepozna?

Prepoznaje i suptilne osmehe.

It does recognize subtle smiles as well.

Vredno smo radili da ovo ostvarimo.

We've worked really hard to make that happen.

Zatim podignute obrve, pokazatelj iznenađenja.

And then eyebrow raised, indicator of surprise.

Brow furrow, which is an indicator of confusion.

Boranje obrva, što je pokazatelj zbunjenosti.

Frown. Yes, perfect.

Mrštenje. Da, savršeno.

So these are all the different action units. There's many more of them.

Ovo su različite akcijske jedinice. Postoji ih još mnogo više.

Ovo je samo skraćena demo verzija.

This is just a slimmed-down demo.

But we call each reading an emotion data point,

Svako čitanje nazivamo tačkom emotivnih podataka

and then they can fire together to portray different emotions.

i one mogu zajedno da rade da iskažu različite emocije.

So on the right side of the demo -- look like you're happy.

Dakle, na desnoj strani demo verzije - izgledaj kao da si srećna.

So that's joy. Joy fires up.

Dakle, to je radost. Radost se povećava.

And then give me a disgust face.

Sad mi pokaži izraz gađenja.

Try to remember what it was like when Zayn left One Direction.

Pokušaj da se setiš kako si se osećala kada je Zejn napustio „One Direction".

(Smeh)

(Laughter)

Da, naboraj nos. Super.

Yeah, wrinkle your nose. Awesome.

And the valence is actually quite negative, so you must have been a big fan.

Valenca je stvarno krajnje negativna, tako da mora da si bila veliki fan.

So valence is how positive or negative an experience is,

Valenca označava koliko je iskustvo pozitivno ili negativno,

and engagement is how expressive she is as well.

a angažman je taj koji označava koliko je ona ekspresivna.

So imagine if Cloe had access to this real-time emotion stream,

Zamislite da Kloi ima pristup ovom emotivnom prenosu u realnom vremenu

and she could share it with anybody she wanted to.

i da može da ga podeli sa kim god želi.

Thank you.

Hvala ti.

(Aplauz)

(Applause)

So, so far, we have amassed 12 billion of these emotion data points.

Dakle, do sada smo nagomilali 12 milijardi ovih tačaka emotivnih podataka.

To je najveća baza emocija u svetu.

It's the largest emotion database in the world.

We've collected it from 2.9 million face videos,

Sakupili smo je kroz 2,9 miliona klipova lica,

people who have agreed to share their emotions with us,

ljudi koji su pristali da podele svoje emocije sa nama,

and from 75 countries around the world.

a to iz 75 zemalja širom sveta.

It's growing every day.

Broj svaki dan raste.

It blows my mind away

Raspamećuje me

that we can now quantify something as personal as our emotions,

to da sada možemo odrediti količinu nečega tako ličnog kao što su emocije

and we can do it at this scale.

i da to možemo obaviti na ovom nivou.

So what have we learned to date?

Dakle, šta smo naučili do sada?

Gender.

Pol.

Our data confirms something that you might suspect.

Naši podaci potvrđuju ono što možda pretpostavljate.

Women are more expressive than men.

Žene su eskpresivnije od muškaraca.

Not only do they smile more, their smiles last longer,

Ne samo da se više osmehuju, njihovi osmesi traju duže.

and we can now really quantify what it is that men and women

Sada stvarno možemo odrediti šta je to na šta muškarci i žene

respond to differently.

reaguju drugačije.

Let's do culture: So in the United States,

Pozabavimo se kulturom. U Sjedinjenim Državama,

women are 40 percent more expressive than men,

žene su 40% ekspresivnije od muškaraca,

but curiously, we don't see any difference in the U.K. between men and women.

ali neobično je to da nema razlike u UK između muškaraca i žena.

(Laughter)

(Smeh)

Age: People who are 50 years and older

Godine. Ljudi koji imaju 50 godina ili stariji od toga

are 25 percent more emotive than younger people.

su 25% emotivniji od mlađih ljudi.

Women in their 20s smile a lot more than men the same age,

Žene u dvadesetima osmehuju se mnogo više od muškaraca istih godina,

perhaps a necessity for dating.

što je možda neophodno pri zabavljanju.

But perhaps what surprised us the most about this data

Ipak, možda najveće iznenađenje u vezi ovih podataka

is that we happen to be expressive all the time,

je to da smo stalno izražajni,

even when we are sitting in front of our devices alone,

čak i kada sedimo sami ispred naših uređaja,

and it's not just when we're watching cat videos on Facebook.

i to ne samo kada gledamo klipove sa mačkama na Fejbuku.

We are expressive when we're emailing, texting, shopping online,

Izražajni smo kada šaljemo i-mejlove, SMS-ove, kupujemo onlajn,

or even doing our taxes.

čak i kada obrađujemo porez.

Where is this data used today?

Gde se ovi podaci koriste danas?

In understanding how we engage with media,

Kada treba da razumemo kako da se angažujemo na mrežama,

so understanding virality and voting behavior;

da razumemo viralnost i ponašanja pri glasanju,

and also empowering or emotion-enabling technology,

kao i da razumemo tehnologije koje osnažuju i omogućuju emocije.

and I want to share some examples that are especially close to my heart.

Želim takođe da podelim i primere koji su mi posebno dragi.

Emotion-enabled wearable glasses can help individuals

Naočare koje omogućuju emocije mogu da pomognu pojedincima

who are visually impaired read the faces of others,

koji imaju oštećen vid da čitaju lica drugih ljudi

and it can help individuals on the autism spectrum interpret emotion,

i mogu da pomognu pojedincima sa autizmom da protumače emocije,

something that they really struggle with.

nešto sa čim stvarno imaju problema.

In education, imagine if your learning apps

U obrazovanju, zamislite kada bi vaše aplikacije za učenje

sense that you're confused and slow down,

mogle da osete kada ste zbunjeni i uspore,

or that you're bored, so it's sped up,

ili kada vam je dosadno i ubrzaju,

just like a great teacher would in a classroom.

kao što bi dobar profesor uradio u učionici.

Šta bi bilo ako bi vaš ručni sat mogao da prati vaše raspoloženje,

What if your wristwatch tracked your mood,

or your car sensed that you're tired,

ili kada bi vaš auto mogao da oseti kada ste umorni,

or perhaps your fridge knows that you're stressed,

ili kada bi vaš frižider znao kada ste pod stresom,

so it auto-locks to prevent you from binge eating. (Laughter)

pa se automatski zatvori da vas spreči da se opsesivno prejedate?

Da, i ja bih to volela.

I would like that, yeah.

(Smeh)

What if, when I was in Cambridge,

Šta bi bilo da sam na Kembridžu

I had access to my real-time emotion stream,

imala pristup svom emotivnom prenosu u realnom vremenu,

and I could share that with my family back home in a very natural way,

da sam mogla da ga podelim to sa porodicom kod kuće na prirodan način,

just like I would've if we were all in the same room together?

kao što bih uradila da smo svi u istoj sobi zajedno?

I think five years down the line,

Mislim da će kroz pet godina,

all our devices are going to have an emotion chip,

svi naši uređaji imati emotivni čip

and we won't remember what it was like when we couldn't just frown at our device

i nećemo se sećati kako je bilo

kada nismo mogli samo da se namrštimo uređajima,

and our device would say, "Hmm, you didn't like that, did you?"

a da oni ne kažu: „Hmm, to ti se nije baš svidelo, zar ne?”

Our biggest challenge is that there are so many applications of this technology,

Naš najveći izazov je veliki broj korisnika ove tehnologije.

my team and I realize that we can't build them all ourselves,

Moj tim i ja smo shvatili da je ne možemo napraviti sami,

so we've made this technology available so that other developers

pa smo ovu tehnologiju učinili dostupnom tako da i drugi programeri

can get building and get creative.

mogu da počnu sa građenjem i iskažu kreativnost.

We recognize that there are potential risks

Razumemo da su mogući i rizici

and potential for abuse,

i da je moguća zloupotreba,

but personally, having spent many years doing this,

ali, budući da sam mnogo godina provela u radu na ovome,

I believe that the benefits to humanity

verujem da su prednosti čovečanstva

sa emocionalno inteligentnom tehnologijom

from having emotionally intelligent technology

far outweigh the potential for misuse.

važnije od moguće zloupotrebe.

And I invite you all to be part of the conversation.

Pozivam vas sve da budete deo te rasprave.

The more people who know about this technology,

Što više ljudi zna za ovu tehnologiju,

the more we can all have a voice in how it's being used.

to će više nas imati pravo da odlučuje kako će se ona koristiti.

So as more and more of our lives become digital,

Dakle, kako naši životi postaju sve više i više digitalni,

we are fighting a losing battle trying to curb our usage of devices

sve više gubimo bitku u pokušaju da smanjimo korišćenje ovih uređaja

in order to reclaim our emotions.

da bismo povratili naše emocije.

So what I'm trying to do instead is to bring emotions into our technology

Dakle, ono što ja pokušavam umesto toga je da uvedem emocije u našu tehnologiju

and make our technologies more responsive.

da bi tehnologija bila prijemčivija.

So I want those devices that have separated us

Dakle, želim da nas ovi uređaji koji su nas rastavili ponovo spoje.

to bring us back together.

And by humanizing technology, we have this golden opportunity

Kada damo ljudska svojstva tehnologiji, dobijamo zlatnu priliku

to reimagine how we connect with machines,

da obnovimo način na koji se povezujemo sa mašinama

and therefore, how we, as human beings,

i stoga, kako se mi, kao ljudska bića,

connect with one another.

povezujemo jedni sa drugima.

Thank you.

Hvala vam.

(Aplauz)

(Applause)