TEDWomen 2013
Rupal Patel: Synthetic voices, as unique as fingerprints
Rupal Patel: Sintetični glasovi, tako unikatni kot prstni odtisi
Filmed:
Readability: 3.9
944,754 views
Veliko ljudi s hudimi motnjami govora uporablja računalniške naprave za komunikacijo. A na izbiro imajo le nekaj glasov. Zato ima Stephen Hawking ameriški naglas in zato veliko ljudi uporablja isti glas, kar ima zelo neskladen učinek. Znanstvenica govora Rupal Patel je hotela glede tega nekaj ukreniti in v tem govoru deli svoje delo ustvarjanja unikatnih glasov za ljudi brez glasu.
Rupal Patel - Speech scientist
People relying on synthetic speech use the voice they’re given, not their own. Rupal Patel created the vocaliD project to change that. Full bio
People relying on synthetic speech use the voice they’re given, not their own. Rupal Patel created the vocaliD project to change that. Full bio
Double-click the English transcript below to play the video.
00:12
I'd like to talk today
0
719
1490
Danes bi rada govorila
00:14
about a powerful and fundamental aspect
1
2209
2927
o močnem in bistvenem vidiku
00:17
of who we are: our voice.
2
5136
3598
tega kar smo. O našem glasu.
00:20
Each one of us has a unique voiceprint
3
8734
2746
Vsak od nas ima unikaten odtis glasu
00:23
that reflects our age, our size,
4
11480
2289
ki odseva našo starost, našo velikost,
00:25
even our lifestyle and personality.
5
13769
3237
celo naš stil življenja in osebnost.
00:29
In the words of the poet Longfellow,
6
17006
2142
V besedah pesnika Longfellowa:
00:31
"the human voice is the organ of the soul."
7
19148
3870
"Človeški glas je glasbilo duše."
00:35
As a speech scientist, I'm fascinated
8
23018
2747
Kot znanstvenico govora me navdušuje
00:37
by how the voice is produced,
9
25765
1829
nastanek zvoka
00:39
and I have an idea for how it can be engineered.
10
27594
3658
in imam idejo, kako bi ga lahko naredili.
00:43
That's what I'd like to share with you.
11
31252
2210
To bi rada delila z vami.
00:45
I'm going to start by playing you a sample
12
33462
1814
Začela bom s predvajanjem vzorca glasu,
00:47
of a voice that you may recognize.
13
35276
1871
ki ga boste morda prepoznali.
00:49
(Recording) Stephen Hawking: "I would have thought
14
37147
1304
(Posnetek) Stephen Hawking: Menil bi,
00:50
it was fairly obvious what I meant."
15
38451
2749
da je bilo popolnoma očitno,
kaj sem mislil."
kaj sem mislil."
00:53
Rupal Patel: That was the voice
16
41200
1280
Rupal Patel: To je bil glas
00:54
of Professor Stephen Hawking.
17
42480
2086
profesorja Stephena Hawkinga.
00:56
What you may not know is that same voice
18
44566
3849
Ne veste pa, da bi isti glas lahko
01:00
may also be used by this little girl
19
48415
2478
uporabljala ta deklica,
01:02
who is unable to speak
20
50893
1697
ki ne more govoriti
01:04
because of a neurological condition.
21
52590
2597
zaradi nevrološke okvare.
01:07
In fact, all of these individuals
22
55187
2068
Pravzaprav, vsi ti posamezniki
01:09
may be using the same voice,
23
57255
2012
bi lahko uporabljali ta glas,
01:11
and that's because there's
only a few options available.
only a few options available.
24
59267
3557
ker je na voljo zelo malo možnosti.
01:14
In the U.S. alone, there are 2.5 million Americans
25
62824
4317
Samo v ZDA,
je okrog 2,5 milijona Američanov
je okrog 2,5 milijona Američanov
01:19
who are unable to speak,
26
67141
1610
ki ne morejo govoriti,
01:20
and many of whom use computerized devices
27
68751
2622
in mnogi izmed njih
uporabljajo računalniške naprave
uporabljajo računalniške naprave
01:23
to communicate.
28
71373
1522
za komunikacijo.
01:24
Now that's millions of people worldwide
29
72895
3479
To so milijoni ljudi po vsem svetu,
01:28
who are using generic voices,
30
76374
1652
ki uporabljajo generični glas
01:30
including Professor Hawking,
31
78026
1446
kot profesor Hawking,
01:31
who uses an American-accented voice.
32
79472
4833
ki uporablja glas z ameriškim naglasom.
01:36
This lack of individuation of the synthetic voice
33
84305
3328
To pomanjkanje individualnosti
sintetičnega glasu
sintetičnega glasu
01:39
really hit home
34
87633
1416
me je res zadelo,
01:41
when I was at an assistive technology conference
35
89049
2472
ko sem bila na konferenci
tehnologije za pomoč
tehnologije za pomoč
01:43
a few years ago,
36
91521
1850
pred nekaj leti,
01:45
and I recall walking into an exhibit hall
37
93371
3604
in spominjam se,
da sem stopila v razstavno dvorano
da sem stopila v razstavno dvorano
01:48
and seeing a little girl and a grown man
38
96975
3044
in videla, kako se deklica
in odrasel moški
in odrasel moški
01:52
having a conversation using their devices,
39
100019
2916
pogovarjata s svojima napravama,
01:54
different devices, but the same voice.
40
102935
4284
različne naprave, a enak glas.
01:59
And I looked around and I saw this happening
41
107219
1909
Pogledala sem okrog in to se je dogajalo
02:01
all around me, literally hundreds of individuals
42
109128
4190
vse okrog mene, na stotine posameznikov
02:05
using a handful of voices,
43
113318
2738
je uporabljalo samo nekaj glasov,
02:08
voices that didn't fit their bodies
44
116056
3091
glasov, ki se niso skladali
z njihovimi telesi
z njihovimi telesi
02:11
or their personalities.
45
119147
2082
ali osebnostmi.
02:13
We wouldn't dream of fitting a little girl
46
121229
2727
Niti v sanjah ne bi deklici namestili
02:15
with the prosthetic limb of a grown man.
47
123956
3396
proteze uda za odraslega moškega.
02:19
So why then the same prosthetic voice?
48
127352
3304
Zakaj torej enak protetični glas?
02:22
It really struck me,
49
130656
1291
Res me je zadelo,
02:23
and I wanted to do something about this.
50
131947
3151
in nekaj sem želela storiti glede tega.
02:27
I'm going to play you now a sample
51
135098
1953
Sedaj vam bom predvajala vzorec
02:29
of someone who has, two people actually,
52
137051
3288
nekoga, ki ima, dva človeka pravzaprav,
02:32
who have severe speech disorders.
53
140339
1768
ki imata hude motnje govora.
02:34
I want you to take a listen to how they sound.
54
142107
3230
Želim, da slišite, kako zvenita.
02:37
They're saying the same utterance.
55
145337
2357
Izgovarjata isti zvok.
02:39
(First voice)
56
147694
2432
(Prvi glas)
02:42
(Second voice)
57
150126
3617
(Drugi glas)
02:45
You probably didn't understand what they said,
58
153743
2412
Najbrž niste razumeli, kaj sta rekla,
02:48
but I hope that you heard
59
156155
1854
ampak upam, da ste slišali
02:50
their unique vocal identities.
60
158009
4283
njuni unikatni glasovni identiteti.
02:54
So what I wanted to do next is,
61
162292
2813
Potem sem hotela
02:57
I wanted to find out how we could harness
62
165105
2384
izvedeti, kako bi lahko izkoristili
02:59
these residual vocal abilities
63
167489
1821
te preostale vokalne zmožnosti
03:01
and build a technology
64
169310
2016
in zgradili tehnologijo,
03:03
that could be customized for them,
65
171326
2143
ki bi jim bila prilagojena,
03:05
voices that could be customized for them.
66
173469
2429
glasovi, ki bi jima bili prilagojeni.
03:07
So I reached out to my collaborator, Tim Bunnell.
67
175898
2685
Povezala sem se s svojim sodelavcem,
Timom Bunnelom.
Timom Bunnelom.
03:10
Dr. Bunnell is an expert in speech synthesis,
68
178583
3063
Dr. Tim Bunnell je strokovnjak
za sintezo govora,
za sintezo govora,
03:13
and what he'd been doing is building
69
181646
2033
in on ustvarja
03:15
personalized voices for people
70
183679
1881
personalizirane glasove za ljudi
03:17
by putting together
71
185560
2097
s sestavljanjem
03:19
pre-recorded samples of their voice
72
187657
2150
prej posnetih vzorcev njihovega glasu
03:21
and reconstructing a voice for them.
73
189807
2879
in rekonstrukcijo glasu zanje.
03:24
These are people who had lost their voice
74
192686
1712
To so ljudje, ki so izgubili svoj glas
03:26
later in life.
75
194398
1911
kasneje v življenju.
03:28
We didn't have the luxury
76
196309
1394
Mi nismo imeli te sreče,
03:29
of pre-recorded samples of speech
77
197703
1774
da bi imeli prej posnete vzorce govora
03:31
for those born with speech disorder.
78
199477
2292
ljudi, ki so rojeni z motnjo govora.
03:33
But I thought, there had to be a way
79
201769
2537
Ampak pomislila sem,
da gotovo obstaja način,
da gotovo obstaja način,
03:36
to reverse engineer a voice
80
204306
1944
da z obratnim inženiringom dobimo glas
03:38
from whatever little is left over.
81
206250
2291
iz tistega, kar je še ostalo.
03:40
So we decided to do exactly that.
82
208541
2714
Točno to smo se odločili storiti.
03:43
We set out with a little bit of funding
from the National Science Foundation,
from the National Science Foundation,
83
211255
3403
Začeli smo z malo financiranja
Nacionalne znanstvene fundacije,
Nacionalne znanstvene fundacije,
03:46
to create custom-crafted voices that captured
84
214658
3565
da bi ustvarili po meri
narejene glasove, ki bi ujeli
narejene glasove, ki bi ujeli
03:50
their unique vocal identities.
85
218223
1536
njihove edinstvene glasovne lastnosti.
03:51
We call this project VocaliD, or vocal I.D.,
86
219759
3203
Temu projektu pravimo VocaliD,
ali vocal I.D.,
ali vocal I.D.,
03:54
for vocal identity.
87
222962
2033
vokalna identiteta.
03:56
Now before I get into the details of how
88
224995
2674
Preden grem v podrobnosti
03:59
the voice is made and let you listen to it,
89
227669
2048
izdelave glasu in vam ga pustim poslušati,
04:01
I need to give you a real quick
speech science lesson. Okay?
speech science lesson. Okay?
90
229717
3350
vam moram res hitro razložiti
osnove znanosti glasu, v redu?
osnove znanosti glasu, v redu?
04:05
So first, we know that the voice is changing
91
233067
3159
Torej, vemo, da se glas dramatično
04:08
dramatically over the course of development.
92
236226
2854
spreminja med razvojem.
04:11
Children sound different from teens
93
239080
2090
Otroci zvenijo drugače od najstnikov,
04:13
who sound different from adults.
94
241170
1463
ki zvenijo drugače kot odrasli.
04:14
We've all experienced this.
95
242633
2642
Vsi smo to izkusili.
04:17
Fact number two is that speech
96
245275
3363
Drugo dejstvo je, da je govor
04:20
is a combination of the source,
97
248638
2553
kombinacija vira,
04:23
which is the vibrations generated by your voice box,
98
251191
3479
to so vibracije, ki jih proizvaja
vaš glasovni aparat,
vaš glasovni aparat,
04:26
which are then pushed through
99
254670
1939
ki jih potem potisnemo
04:28
the rest of the vocal tract.
100
256609
2437
skozi preostali vokalni trakt.
04:31
These are the chambers of your head and neck
101
259046
2484
To so votline v vaši glavi in vratu,
04:33
that vibrate,
102
261530
1239
ki vibrirajo,
04:34
and they actually filter that source sound
103
262769
2110
in dejansko filtrirajo ta izvorni zvok,
04:36
to produce consonants and vowels.
104
264879
2537
da ustvarijo soglasnike in samoglasnike.
04:39
So the combination of source and filter
105
267416
3860
S kombinacija izvora in filtra
04:43
is how we produce speech.
106
271276
2630
mi proizvajamo zvok.
04:45
And that happens in one individual.
107
273906
3026
In to se zgodi v enem posamezniku.
04:48
Now I told you earlier that I'd spent
108
276932
2626
Prej sem vam povedala, da sem preživela
04:51
a good part of my career
109
279558
2025
dober del svoje kariere,
04:53
understanding and studying
110
281583
2453
da sem preučevala in skušala razumeti
04:56
the source characteristics of people
111
284036
1958
izvorne lastnosti ljudi
04:57
with severe speech disorder,
112
285994
2301
s hudimi motnjami govora,
05:00
and what I've found
113
288295
1465
in odkrila sem,
05:01
is that even though their filters were impaired,
114
289760
3366
da čeprav so bili
njihovi filtri poškodovani,
njihovi filtri poškodovani,
05:05
they were able to modulate their source:
115
293126
2961
so lahko uravnavali svoj vir:
05:08
the pitch, the loudness, the tempo of their voice.
116
296087
3262
višino, glasnost, tempo svojega glasu.
05:11
These are called prosody, and
I've been documenting for years
I've been documenting for years
117
299349
3368
To se imenuje prozodija,
in že leta dokumentiram,
in že leta dokumentiram,
05:14
that the prosodic abilities of these individuals
118
302717
2277
da so prozodične sposobnosti
teh posameznikov
teh posameznikov
05:16
are preserved.
119
304994
1575
ohranjene.
05:18
So when I realized that those same cues
120
306569
4087
Ko sem spoznala, da so te iste iztočnice
05:22
are also important for speaker identity,
121
310656
2769
prav tako pomembne za identiteto govorca,
05:25
I had this idea.
122
313425
2015
sem dobila idejo.
05:27
Why don't we take the source
123
315440
2516
Zakaj ne vzamemo vira
05:29
from the person we want the voice to sound like,
124
317956
2213
nekoga, čigar glas si želimo,
05:32
because it's preserved,
125
320169
1463
ker je ohranjen,
05:33
and borrow the filter
126
321632
2135
in si filter sposodimo
05:35
from someone about the same age and size,
127
323767
3229
od nekoga istih let in velikosti,
05:39
because they can articulate speech,
128
327011
2407
ker lahko artikulirajo svoj govor,
05:41
and then mix them?
129
329418
1791
in jih potem pomešamo?
05:43
Because when we mix them,
130
331209
1787
Ker ko jih pomešamo,
05:44
we can get a voice that's as clear
131
332996
1698
dobimo glas, ki je tako jasen
05:46
as our surrogate talker --
132
334694
1754
kot naš govornik darovalec -
05:48
that's the person we borrowed the filter from—
133
336448
2595
to je oseba, pri kateri
smo si izposodili filter -
smo si izposodili filter -
05:51
and is similar in identity to our target talker.
134
339043
4649
in je v identiteti podoben našemu govorcu.
05:55
It's that simple.
135
343692
1427
Tako preprosto je.
05:57
That's the science behind what we're doing.
136
345119
2934
To je znanost za našim delom.
06:00
So once you have that in mind,
137
348053
3533
Ko imaš enkrat to v mislih,
06:03
how do you go about building this voice?
138
351586
2258
kako potem zgradiš ta glas?
06:05
Well, you have to find someone
139
353844
1480
No, najti moraš nekoga,
06:07
who is willing to be a surrogate.
140
355324
2400
ki je pripravljen biti darovalec.
06:09
It's not such an ominous thing.
141
357724
2264
To ni tako zlovešča stvar.
06:11
Being a surrogate donor
142
359988
1523
Biti darovalec
06:13
only requires you to say a few hundred
143
361511
2788
zahteva samo izgovorjavo nekaj sto
06:16
to a few thousand utterances.
144
364299
2242
do nekaj tisoč glasov.
06:18
The process goes something like this.
145
366541
2003
To proces poteka nekako takole.
06:20
(Video) Voice: Things happen in pairs.
146
368544
2190
(Video) Glas: Stvari se dogajajo v parih.
06:22
I love to sleep.
147
370734
1925
Rad spim.
06:24
The sky is blue without clouds.
148
372659
3882
Nebo je modro brez oblakov.
06:28
RP: Now she's going to go on like this
149
376541
2002
RP: Takole bo šla naprej
06:30
for about three to four hours,
150
378543
1919
približno tri do štiri ure,
06:32
and the idea is not for her to say everything
151
380462
3005
in ideja ni, da izreče vse,
06:35
that the target is going to want to say,
152
383467
2045
kar bo hotela izreči prejemnica,
06:37
but the idea is to cover all the different combinations
153
385512
3395
ampak je ideja, da pokrijemo
vse različne kombinacije
vse različne kombinacije
06:40
of the sounds that occur in the language.
154
388907
3271
zvokov, ki se lahko pojavijo v jeziku.
06:44
The more speech you have,
155
392178
1638
Več govora kot imaš,
06:45
the better sounding voice you're going to have.
156
393816
2305
lepše zveneč glas boš imel.
06:48
Once you have those recordings,
157
396121
1673
Ko imamo te posnetke,
06:49
what we need to do
158
397794
1413
jih moramo
06:51
is we have to parse these recordings
159
399207
2718
razčleniti
06:53
into little snippets of speech,
160
401925
2449
v majhne drobce govora,
06:56
one- or two-sound combinations,
161
404374
2337
kombinacije enega ali dveh zvokov,
06:58
sometimes even whole words
162
406711
1883
včasih celo cele besede
07:00
that start populating a dataset or a database.
163
408594
4516
da začnemo zapolnjevati
set podatkov ali podatkovno bazo.
set podatkov ali podatkovno bazo.
07:05
We're going to call this database a voice bank.
164
413110
3717
Tej bazi podatkov
bomo rekli glasovna banka.
bomo rekli glasovna banka.
07:08
Now the power of the voice bank
165
416827
2096
Moč glasovne banke
07:10
is that from this voice bank,
166
418923
2014
je v tem, da iz nje
07:12
we can now say any new utterance,
167
420937
2011
lahko izgovorimo karkoli,
07:14
like, "I love chocolate" --
168
422948
1424
recimo: "Rada imam čokolado" -
07:16
everyone needs to be able to say that—
169
424372
1739
vsakdo mora biti sposoben izreči to -
07:18
fish through that database
170
426111
1831
pobrskamo po bazi podatkov
07:19
and find all the segments necessary
171
427942
1940
in najdemo vse potrebne segmente
07:21
to say that utterance.
172
429882
1929
za izgovorjavo tega.
07:23
(Video) Voice: I love chocolate.
173
431811
1789
(Video) Glas: Rada imam čokolado.
07:25
RP: So that's speech synthesis.
174
433600
1391
RP: To je sinteza govora.
07:26
It's called concatenative synthesis,
and that's what we're using.
and that's what we're using.
175
434991
2573
To je konkatenativna sinteza,
in to mi uporabljamo.
in to mi uporabljamo.
07:29
That's not the novel part.
176
437564
1533
To ni nekaj novega.
07:31
What's novel is how we make it sound
177
439097
2221
Nov del je, kako naredimo, da zveni
07:33
like this young woman.
178
441318
1457
kot ta mlada ženska.
07:34
This is Samantha.
179
442775
1524
To je Samantha.
07:36
I met her when she was nine,
180
444299
2346
Spoznala sem jo, ko je imela 9 let,
07:38
and since then, my team and I
181
446645
1897
in od takrat ji z ekipo
07:40
have been trying to build her a personalized voice.
182
448542
2714
poskušamo zgraditi njen osebni glas.
07:43
We first had to find a surrogate donor,
183
451256
3099
Najprej smo morali najti
nadomestnega donorja,
nadomestnega donorja,
07:46
and then we had to have Samantha
184
454355
1818
potem je morala Samantha izgovoriti
07:48
produce some utterances.
185
456173
1929
nekaj zlogov.
07:50
What she can produce are mostly vowel-like sounds,
186
458102
2379
Lahko proizvaja predvsem samoglasnike,
07:52
but that's enough for us to extract
187
460481
2479
a to je bilo za nas dovolj,
da smo lahko pridobili
da smo lahko pridobili
07:54
her source characteristics.
188
462960
2285
njene značilnosti.
07:57
What happens next is best described
189
465245
3271
Kar se zgodi potem, najlepše opiše
08:00
by my daughter's analogy. She's six.
190
468516
2767
analogija moje hčere. Šest let ima.
08:03
She calls it mixing colors to paint voices.
191
471283
5422
Temu pravi, da mešamo barve,
za risanje glasov.
za risanje glasov.
08:08
It's beautiful. It's exactly that.
192
476705
2555
Prelepo je. Točno tako je.
08:11
Samantha's voice is like a concentrated sample
193
479260
2860
Samanthin glas je kot koncentriran vzorec
08:14
of red food dye which we can infuse
194
482120
2609
rdečega barvila za hrano, ki ga lahko damo
08:16
into the recordings of her surrogate
195
484729
2540
v posnetke njenega darovalca,
08:19
to get a pink voice just like this.
196
487269
4387
da dobimo tak rožnat glas.
08:23
(Video) Samantha: Aaaaaah.
197
491656
4491
(video) Samantha: Aaaaah.
08:28
RP: So now, Samantha can say this.
198
496147
2808
RP: Sedaj, lahko Samantha reče tole.
08:30
(Video) Samantha: This voice is only for me.
199
498955
3069
(Video) Samantha: Ta glas je samo zame.
08:34
I can't wait to use my new voice with my friends.
200
502024
6305
Komaj čakam, da uporabim
svoj glas s svojimi prijatelji.
svoj glas s svojimi prijatelji.
08:40
RP: Thank you. (Applause)
201
508329
6417
RP: Hvala. (Aplavz)
08:46
I'll never forget the gentle smile
202
514746
2333
Nikoli ne bom pozabila nežnega nasmeška,
08:49
that spread across her face
203
517079
1902
ki se je razširil čez njen obraz,
08:50
when she heard that voice for the first time.
204
518981
3649
ko je prvič slišala ta glas.
08:54
Now there's millions of people
205
522630
1882
Tu so milijoni ljudi
08:56
around the world like Samantha, millions,
206
524512
2833
po celem svetu,
ki so kot Samantha, milijoni,
ki so kot Samantha, milijoni,
08:59
and we've only begun to scratch the surface.
207
527345
3440
in šele začeli smo.
09:02
What we've done so far is we have
208
530785
1642
Sedaj imamo
09:04
a few surrogate talkers from around the U.S.
209
532427
3859
nekaj nadomestnih govorcev
po celih Združenih državah Amerike,
po celih Združenih državah Amerike,
09:08
who have donated their voices,
210
536286
1507
ki so darovali svoj glas,
09:09
and we have been using those
211
537793
1928
in mi smo jih uporabljali,
09:11
to build our first few personalized voices.
212
539721
4472
da smo zgradili naše prve
personalizirane glasove.
personalizirane glasove.
09:16
But there's so much more work to be done.
213
544193
1756
A veliko dela je še pred nami.
09:17
For Samantha, her surrogate
214
545949
2188
Za Samantho, njen donor
09:20
came from somewhere in the Midwest, a stranger
215
548137
3046
prihaja nekje iz srednjega zahoda, tujec,
09:23
who gave her the gift of voice.
216
551183
3841
ki ji je dal dar glasu.
09:27
And as a scientist, I'm so excited
217
555024
2153
Kot znanstvenica, sem tako vznemirjena,
09:29
to take this work out of the laboratory
218
557177
1935
da smo vzeli to delo iz laboratorija
09:31
and finally into the real world
219
559112
1800
v realni svet,
09:32
so it can have real-world impact.
220
560912
3165
kjer ima lahko realen vpliv.
09:36
What I want to share with you next
221
564077
1582
Z vami bi rada delila,
09:37
is how I envision taking this work
222
565659
2175
kako bi popeljala to delo
09:39
to that next level.
223
567834
2711
na naslednjo raven.
09:42
I imagine a whole world of surrogate donors
224
570545
3887
Predstavljam si
cel svet nadomestnih darovalcev,
cel svet nadomestnih darovalcev,
09:46
from all walks of life, different sizes, different ages,
225
574432
3260
vseh vrst, različnih velikosti,
različne starosti,
različne starosti,
09:49
coming together in this voice drive
226
577692
3058
kako se srečujejo v tej bazi glasov,
09:52
to give people voices
227
580750
2270
da bi dali ljudem glasove,
09:55
that are as colorful as their personalities.
228
583020
3799
ki so tako barviti kot njihove osebnosti.
09:58
To do that as a first step,
229
586819
2300
Prvi korak je bila
10:01
we've put together this website, VocaliD.org,
230
589119
3275
postavitev spletne strani, VocaliD.org,
10:04
as a way to bring together those
231
592394
1624
da bi zbrali tiste,
10:06
who want to join us as voice donors,
232
594018
2675
ki bi se nam radi pridružili
kot darovalci glasov,
kot darovalci glasov,
10:08
as expertise donors,
233
596693
1772
kot izkušeni darovalci,
10:10
in whatever way to make this vision a reality.
234
598465
5339
na kakršenkoli način,
da bi ta načrt postal resničnost.
da bi ta načrt postal resničnost.
10:15
They say that giving blood can save lives.
235
603804
4153
Pravijo, da darovanje krvi
rešuje življenja.
rešuje življenja.
10:19
Well, giving your voice can change lives.
236
607957
4982
No, če posodite svoj glas,
lahko spremenite življenja.
lahko spremenite življenja.
10:24
All we need is a few hours of speech
237
612939
3050
Vse, kar potrebujemo, je nekaj ur govora
10:27
from our surrogate talker,
238
615989
1491
našega nadomestnega govorca
10:29
and as little as a vowel from our target talker,
239
617480
4733
in samo samostalnik našega govorca,
10:34
to create a unique vocal identity.
240
622213
3711
da ustvarimo unikatno glasovno identiteto.
10:37
So that's the science behind what we're doing.
241
625924
2626
To je znanost za našim delom.
10:40
I want to end by circling back to the human side
242
628550
4455
Zaključila bi s človeško platjo,
10:45
that is really the inspiration for this work.
243
633005
4102
ki je resnični navdih za to delo.
10:49
About five years ago, we built our very first voice
244
637107
3699
Pred petimi leti,
smo zgradili naš prvi glas
smo zgradili naš prvi glas
10:52
for a little boy named William.
245
640806
2501
za majhnega fanta po imenu William.
10:55
When his mom first heard this voice,
246
643307
2357
Ko je njegova mama prvič slišala ta glas,
10:57
she said, "This is what William
247
645664
2345
je rekla: "Tako bi William zvenel,
11:00
would have sounded like
248
648009
1546
11:01
had he been able to speak."
249
649555
2449
če bi lahko govoril."
11:04
And then I saw William typing a message
250
652004
2418
Videla sem Williama, kako tipka sporočilo
11:06
on his device.
251
654422
1362
na svoji napravi.
11:07
I wondered, what was he thinking?
252
655784
3293
Spraševala sem se, le o čem razmišlja?
11:11
Imagine carrying around someone else's voice
253
659077
3590
Predstavljajte si,
da nosite glas nekoga drugega
da nosite glas nekoga drugega
11:14
for nine years
254
662667
2193
devet let
11:16
and finally finding your own voice.
255
664860
4844
in končno najdete svoj glas.
11:21
Imagine that.
256
669704
1377
Predstavljajte si.
11:23
This is what William said:
257
671081
2797
To je rekel William:
11:25
"Never heard me before."
258
673878
4463
"Sebe pa še nisem slišal."
11:32
Thank you.
259
680417
1619
Hvala.
11:34
(Applause)
260
682036
4724
(Aplavz)
ABOUT THE SPEAKER
Rupal Patel - Speech scientistPeople relying on synthetic speech use the voice they’re given, not their own. Rupal Patel created the vocaliD project to change that.
Why you should listen
Northeastern University computer science professor Rupal Patel looks for ways to give voice to the voiceless. As founder and director of the Communication Analysis and Design Laboratory (CadLab), she developed a technology that combines real human voices with the characteristics of individual speech patterns. The result is VocaliD, an innovation that gives people who can't speak the ability to communicate in a voice all their own.
"There's nothing better than seeing the person who's actually going to use it, seeing their reaction, seeing their smile," says Patel.
Rupal Patel | Speaker | TED.com