TEDWomen 2013
Rupal Patel: Synthetic voices, as unique as fingerprints
卢帕尔·帕特尔: 和指纹一样独特的合成声音
Filmed:
Readability: 3.9
944,754 views
很多有着严重言语障碍的人都用一个电脑化的设备进行交流。但是，他们的声音选择却非常有限。这就是为什么史蒂芬·霍金教授有美国口音，为什么很多人有着同样的声音，不符合他们身份的声音。语音学家卢帕尔·帕特尔想对此做点什么。在这个很棒的演讲中，她跟我们分享了她如何为那些没有声音的人打造独特的声音的过程。
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
我今天要和大家讲述的是
00:14
about a powerful and fundamental aspect
1
2209
2927
关于我们自身的一个非常强大
00:17
of who we are: our voice.
2
5136
3598
非常重要的方面：我们的声音，
00:20
Each one of us has a unique voiceprint
3
8734
2746
每一个人的声音都带有独特的标记，
00:23
that reflects our age, our size,
4
11480
2289
这个声音的标记能反映出我们的年龄，我们的胖瘦高矮，
00:25
even our lifestyle and personality.
5
13769
3237
甚至是我们的生活方式和性格。
00:29
In the words of the poet Longfellow,
6
17006
2142
用诗人朗费罗的话来说，
00:31
"the human voice is the organ of the soul."
7
19148
3870
“人类的声音是灵魂的重要器官。”
00:35
As a speech scientist, I'm fascinated
8
23018
2747
身为一个语音科学家，我非常热衷于研究
00:37
by how the voice is produced,
9
25765
1829
声音的产生，
00:39
and I have an idea for how it can be engineered.
10
27594
3658
而且我有一个如何制造声音的想法。
00:43
That's what I'd like to share with you.
11
31252
2210
这就是我今天想和大家分享的东西。
00:45
I'm going to start by playing you a sample
12
33462
1814
首先，我想为大家播放一个声音样本，
00:47
of a voice that you may recognize.
13
35276
1871
这个声音你们可能听过。
00:49
(Recording) Stephen Hawking: "I would have thought
14
37147
1304
（录音）史蒂芬·霍金：“我本来以为，
00:50
it was fairly obvious what I meant."
15
38451
2749
我想说的意思很显而易见。”
00:53
Rupal Patel: That was the voice
16
41200
1280
卢帕尔·帕特尔：那是
00:54
of Professor Stephen Hawking.
17
42480
2086
史蒂芬·霍金教授的声音。
00:56
What you may not know is that same voice
18
44566
3849
你们可能不知道的是，同样的声音
01:00
may also be used by this little girl
19
48415
2478
也被用于这个小女孩身上，
01:02
who is unable to speak
20
50893
1697
她因为大脑神经系统缺陷
01:04
because of a neurological condition.
21
52590
2597
而不能讲话。
01:07
In fact, all of these individuals
22
55187
2068
事实上，很多不能说话的人
01:09
may be using the same voice,
23
57255
2012
都可能在使用同样的声音
01:11
and that's because there's
only a few options available.
only a few options available.
24
59267
3557
那是因为可以使用的声音样本太少了。
01:14
In the U.S. alone, there are 2.5 million Americans
25
62824
4317
单单在美国，就有250万人
01:19
who are unable to speak,
26
67141
1610
不能说话，
01:20
and many of whom use computerized devices
27
68751
2622
而且在这些人中很多都是使用电脑化的设备
01:23
to communicate.
28
71373
1522
进行交流。
01:24
Now that's millions of people worldwide
29
72895
3479
也就是全世界数百万的人
01:28
who are using generic voices,
30
76374
1652
都在使用一些毫无个性的声音，
01:30
including Professor Hawking,
31
78026
1446
其中就包括史蒂芬·霍金教授，
01:31
who uses an American-accented voice.
32
79472
4833
他使用的声音是带有美国口音的。
01:36
This lack of individuation of the synthetic voice
33
84305
3328
我真正开始意识到
01:39
really hit home
34
87633
1416
合成声音缺乏个性
01:41
when I was at an assistive technology conference
35
89049
2472
是我在几年前参加一个
01:43
a few years ago,
36
91521
1850
辅助技术会议的时候，
01:45
and I recall walking into an exhibit hall
37
93371
3604
我记得走进一个展厅，
01:48
and seeing a little girl and a grown man
38
96975
3044
看到一个小女孩和一个成年男子
01:52
having a conversation using their devices,
39
100019
2916
正在用他们的设备进行对话，
01:54
different devices, but the same voice.
40
102935
4284
不同的设备，却是同样的声音。
01:59
And I looked around and I saw this happening
41
107219
1909
我看向四周，发现身边这种情况很多，
02:01
all around me, literally hundreds of individuals
42
109128
4190
几乎是上百个人
02:05
using a handful of voices,
43
113318
2738
却只用着为数不多的几种声音，
02:08
voices that didn't fit their bodies
44
116056
3091
这些声音跟他们的身体特征
02:11
or their personalities.
45
119147
2082
和性格都很不匹配。
02:13
We wouldn't dream of fitting a little girl
46
121229
2727
我们肯定做梦也不会想到把一个成年男子的假肢
02:15
with the prosthetic limb of a grown man.
47
123956
3396
装在一个小女孩身上。
02:19
So why then the same prosthetic voice?
48
127352
3304
那为什么他们要用同样的合成声音呢？
02:22
It really struck me,
49
130656
1291
这深深的触动了我，
02:23
and I wanted to do something about this.
50
131947
3151
我想做些什么。
02:27
I'm going to play you now a sample
51
135098
1953
现在我想为大家播放一个人的录音——
02:29
of someone who has, two people actually,
52
137051
3288
不对，其实是两个人，
02:32
who have severe speech disorders.
53
140339
1768
他们都有很严重的言语障碍。
02:34
I want you to take a listen to how they sound.
54
142107
3230
我想让大家听听他们的声音。
02:37
They're saying the same utterance.
55
145337
2357
他们在发出同样一个音。
02:39
(First voice)
56
147694
2432
（第一个声音）
02:42
(Second voice)
57
150126
3617
（第二个声音）
02:45
You probably didn't understand what they said,
58
153743
2412
大家可能并不明白他们说了什么，
02:48
but I hope that you heard
59
156155
1854
但我希望大家听到了
02:50
their unique vocal identities.
60
158009
4283
他们独特的声音标志。
02:54
So what I wanted to do next is,
61
162292
2813
所以接下来我想要做的事情就是，
02:57
I wanted to find out how we could harness
62
165105
2384
我想要找出如何可以利用
02:59
these residual vocal abilities
63
167489
1821
他们残留的发声能力，
03:01
and build a technology
64
169310
2016
并发明一项技术，
03:03
that could be customized for them,
65
171326
2143
这项技术能为他们创造出个性化的声音，
03:05
voices that could be customized for them.
66
173469
2429
就是专门为他们定制的声音。
03:07
So I reached out to my collaborator, Tim Bunnell.
67
175898
2685
所以我联系了我的合作伙伴，蒂姆·邦内尔。
03:10
Dr. Bunnell is an expert in speech synthesis,
68
178583
3063
邦内尔博士是一位语言合成方面的专家，
03:13
and what he'd been doing is building
69
181646
2033
他一直在为需要帮助的人合成
03:15
personalized voices for people
70
183679
1881
个性化的声音，
03:17
by putting together
71
185560
2097
他把这些人
03:19
pre-recorded samples of their voice
72
187657
2150
预先录制好的声音样本组合在一起，
03:21
and reconstructing a voice for them.
73
189807
2879
并重新建立他们的声音。
03:24
These are people who had lost their voice
74
192686
1712
这些人都是在人生后来的某个阶段
03:26
later in life.
75
194398
1911
才失去了语言能力。
03:28
We didn't have the luxury
76
196309
1394
可是我们没有
03:29
of pre-recorded samples of speech
77
197703
1774
那些生来就有言语障碍的人的
03:31
for those born with speech disorder.
78
199477
2292
预先录制好的声音样本。
03:33
But I thought, there had to be a way
79
201769
2537
但我想，肯定有一个办法
03:36
to reverse engineer a voice
80
204306
1944
可以利用仅存的不管剩下多少的语言能力
03:38
from whatever little is left over.
81
206250
2291
来逆向重组声音。
03:40
So we decided to do exactly that.
82
208541
2714
于是我们决定去做这样的工作。
03:43
We set out with a little bit of funding
from the National Science Foundation,
from the National Science Foundation,
83
211255
3403
我们从国家科学基金会的一小笔资金开始，
03:46
to create custom-crafted voices that captured
84
214658
3565
努力打造反映了他们的独特声印的
03:50
their unique vocal identities.
85
218223
1536
定制的声音。
03:51
We call this project VocaliD, or vocal I.D.,
86
219759
3203
我们称之为VocaliD计划，即声音ID，
03:54
for vocal identity.
87
222962
2033
用于区别不同的声音。
03:56
Now before I get into the details of how
88
224995
2674
那么，在我开始讲述
03:59
the voice is made and let you listen to it,
89
227669
2048
声音是如何制作的，以及让大家听这些声音之前，
04:01
I need to give you a real quick
speech science lesson. Okay?
speech science lesson. Okay?
90
229717
3350
我需要先给大家上一堂关于语音学的快速入门课，
可以么？
可以么？
04:05
So first, we know that the voice is changing
91
233067
3159
首先，我们知道声音
04:08
dramatically over the course of development.
92
236226
2854
在其发展过程中会发生巨大的改变。
04:11
Children sound different from teens
93
239080
2090
儿童的声音与青少年的声音不同，
04:13
who sound different from adults.
94
241170
1463
而青少年的声音则与成人的声音不同。
04:14
We've all experienced this.
95
242633
2642
我们都经历过这样的改变。
04:17
Fact number two is that speech
96
245275
3363
第二，语音是
04:20
is a combination of the source,
97
248638
2553
声源的组合，
04:23
which is the vibrations generated by your voice box,
98
251191
3479
也就是你的喉部产生的震动
04:26
which are then pushed through
99
254670
1939
通过声道
04:28
the rest of the vocal tract.
100
256609
2437
传出来。
04:31
These are the chambers of your head and neck
101
259046
2484
这些是你的头部和颈部
04:33
that vibrate,
102
261530
1239
会震动的腔室，
04:34
and they actually filter that source sound
103
262769
2110
他们会过滤声源
04:36
to produce consonants and vowels.
104
264879
2537
并产生辅音和元音。
04:39
So the combination of source and filter
105
267416
3860
所以声源和过滤器的组合
04:43
is how we produce speech.
106
271276
2630
使得我们能够制造语言。
04:45
And that happens in one individual.
107
273906
3026
而这发生在一个个体身上。
04:48
Now I told you earlier that I'd spent
108
276932
2626
早先我告诉过你们
04:51
a good part of my career
109
279558
2025
我花了我职业生涯中的很大一部分时间
04:53
understanding and studying
110
281583
2453
来了解和学习
04:56
the source characteristics of people
111
284036
1958
那些有着严重言语障碍的人的
04:57
with severe speech disorder,
112
285994
2301
声源的特征，
05:00
and what I've found
113
288295
1465
我发现
05:01
is that even though their filters were impaired,
114
289760
3366
虽然他们的过滤器受损，
05:05
they were able to modulate their source:
115
293126
2961
他们仍然能够控制他们的声源，
05:08
the pitch, the loudness, the tempo of their voice.
116
296087
3262
包括音高、响度和声音的节奏。
05:11
These are called prosody, and
I've been documenting for years
I've been documenting for years
117
299349
3368
这些我们称这些为韵律，而我多年的记录表明
05:14
that the prosodic abilities of these individuals
118
302717
2277
这些人的韵律能力
05:16
are preserved.
119
304994
1575
被保留了下来。
05:18
So when I realized that those same cues
120
306569
4087
所以当我意识到这些同样的线索
05:22
are also important for speaker identity,
121
310656
2769
对讲者身份也是非常重要的时候，
05:25
I had this idea.
122
313425
2015
我有了这样一个想法。
05:27
Why don't we take the source
123
315440
2516
为什么不利用那些
05:29
from the person we want the voice to sound like,
124
317956
2213
我们希望听到的声音的声源，
05:32
because it's preserved,
125
320169
1463
因为这个声源是好的，
05:33
and borrow the filter
126
321632
2135
再借助一个
05:35
from someone about the same age and size,
127
323767
3229
差不多年龄和体型的人的过滤器，
05:39
because they can articulate speech,
128
327011
2407
因为他们可以清晰地发声，
05:41
and then mix them?
129
329418
1791
然后把他们组合在一起？
05:43
Because when we mix them,
130
331209
1787
因为当我们把它们组合在一起的时候，
05:44
we can get a voice that's as clear
131
332996
1698
我们就可以获得一个
05:46
as our surrogate talker --
132
334694
1754
像代理说话者一样清晰的声音，
05:48
that's the person we borrowed the filter from—
133
336448
2595
代理说话者就是我们向其借了过滤器的那个人，
05:51
and is similar in identity to our target talker.
134
339043
4649
而这个声音又跟我们的目标说话者的身份一致。
05:55
It's that simple.
135
343692
1427
就这么简单。
05:57
That's the science behind what we're doing.
136
345119
2934
这就是我们在做的研究背后的科学。
06:00
So once you have that in mind,
137
348053
3533
有了这样的想法以后，
06:03
how do you go about building this voice?
138
351586
2258
我们又该如何真正去打造这样的声音呢？
06:05
Well, you have to find someone
139
353844
1480
嗯，你必须找到
06:07
who is willing to be a surrogate.
140
355324
2400
愿意做代理说话者的人。
06:09
It's not such an ominous thing.
141
357724
2264
这并不是什么有着不祥之兆的事情。
06:11
Being a surrogate donor
142
359988
1523
作为一个代理说话者，
06:13
only requires you to say a few hundred
143
361511
2788
你只需要说上几百个
06:16
to a few thousand utterances.
144
364299
2242
到几千个话语。
06:18
The process goes something like this.
145
366541
2003
过程大致是这样的。
06:20
(Video) Voice: Things happen in pairs.
146
368544
2190
（视频）声音：事情成对发生。
06:22
I love to sleep.
147
370734
1925
我爱睡觉。
06:24
The sky is blue without clouds.
148
372659
3882
天空很蓝，无云。
06:28
RP: Now she's going to go on like this
149
376541
2002
卢帕尔·帕特尔：她就这样继续说上
06:30
for about three to four hours,
150
378543
1919
大约三到四个小时，
06:32
and the idea is not for her to say everything
151
380462
3005
当然她并不需要说出
06:35
that the target is going to want to say,
152
383467
2045
目标说话者会说的所有东西，
06:37
but the idea is to cover all the different combinations
153
385512
3395
而只需覆盖到一门语言中的
06:40
of the sounds that occur in the language.
154
388907
3271
所有发音的不同组合。
06:44
The more speech you have,
155
392178
1638
越多的语音样本
06:45
the better sounding voice you're going to have.
156
393816
2305
就意味着越好的声音质量。
06:48
Once you have those recordings,
157
396121
1673
一旦有了这些录音之后，
06:49
what we need to do
158
397794
1413
我们需要做的就是
06:51
is we have to parse these recordings
159
399207
2718
将这些录音
06:53
into little snippets of speech,
160
401925
2449
解析成语音的小片段，
06:56
one- or two-sound combinations,
161
404374
2337
一两个发声的组合，
06:58
sometimes even whole words
162
406711
1883
有的时候甚至整个的词语
07:00
that start populating a dataset or a database.
163
408594
4516
也会出现在数据库里边。
07:05
We're going to call this database a voice bank.
164
413110
3717
我们就将这个数据库称为声音银行。
07:08
Now the power of the voice bank
165
416827
2096
这个声音银行的作用在于：
07:10
is that from this voice bank,
166
418923
2014
基于这个声音银行，
07:12
we can now say any new utterance,
167
420937
2011
我们现在可以说出任何新的话语，
07:14
like, "I love chocolate" --
168
422948
1424
比如：“我爱巧克力”——
07:16
everyone needs to be able to say that—
169
424372
1739
每个人都应该有可以说出这句话的能力——
07:18
fish through that database
170
426111
1831
从这个数据库中寻找
07:19
and find all the segments necessary
171
427942
1940
并找到说这句话需要的
07:21
to say that utterance.
172
429882
1929
所有必要的片段。
07:23
(Video) Voice: I love chocolate.
173
431811
1789
（视频）声音：我爱巧克力。
07:25
RP: So that's speech synthesis.
174
433600
1391
卢帕尔·帕特尔：这就是语音合成。
07:26
It's called concatenative synthesis,
and that's what we're using.
and that's what we're using.
175
434991
2573
这个被称之为衔接合成，而我们用的就是它。
07:29
That's not the novel part.
176
437564
1533
其实这部分并不新奇。
07:31
What's novel is how we make it sound
177
439097
2221
新奇的部分是我们如何制作出听起来
07:33
like this young woman.
178
441318
1457
像是这个年轻女性的声音。
07:34
This is Samantha.
179
442775
1524
这是萨曼莎。
07:36
I met her when she was nine,
180
444299
2346
我第一次见到她的时候，她九岁，
07:38
and since then, my team and I
181
446645
1897
从那时候起，我和我的团队
07:40
have been trying to build her a personalized voice.
182
448542
2714
就一直在努力给她打造一个属于她自己的声音。
07:43
We first had to find a surrogate donor,
183
451256
3099
我们首先要找到一个代理说话者，
07:46
and then we had to have Samantha
184
454355
1818
然后我们让萨曼莎
07:48
produce some utterances.
185
456173
1929
发出一些声音。
07:50
What she can produce are mostly vowel-like sounds,
186
458102
2379
她能做的就是发出一些类似元音的声音，
07:52
but that's enough for us to extract
187
460481
2479
但这对于我们提取她的声源特征
07:54
her source characteristics.
188
462960
2285
已经足够了。
07:57
What happens next is best described
189
465245
3271
接下来发生的事情最好可以
08:00
by my daughter's analogy. She's six.
190
468516
2767
用我女儿的比喻来描述。她六岁。
08:03
She calls it mixing colors to paint voices.
191
471283
5422
她称其为“用不同的颜色画声音”。
08:08
It's beautiful. It's exactly that.
192
476705
2555
美极了。正是这样。
08:11
Samantha's voice is like a concentrated sample
193
479260
2860
萨曼莎的声音就好比是
08:14
of red food dye which we can infuse
194
482120
2609
浓缩的红色食用色素注入了
08:16
into the recordings of her surrogate
195
484729
2540
她的代理说话者的录音里面，
08:19
to get a pink voice just like this.
196
487269
4387
而产生了这样的粉红色的声音。
08:23
(Video) Samantha: Aaaaaah.
197
491656
4491
（视频）萨曼莎：啊……
08:28
RP: So now, Samantha can say this.
198
496147
2808
卢帕尔·帕特尔：那么现在，萨曼莎可以说这样的话。
08:30
(Video) Samantha: This voice is only for me.
199
498955
3069
（视频）萨曼莎：这是只属于我的声音。
08:34
I can't wait to use my new voice with my friends.
200
502024
6305
我迫不及待地想跟我的朋友用我 的新声音交流。
08:40
RP: Thank you. (Applause)
201
508329
6417
卢帕尔·帕特尔：谢谢。（掌声）
08:46
I'll never forget the gentle smile
202
514746
2333
我永远不会忘记
08:49
that spread across her face
203
517079
1902
当她第一次听到自己的声音的时候，
08:50
when she heard that voice for the first time.
204
518981
3649
那个绽放在她脸上的温柔的笑脸。
08:54
Now there's millions of people
205
522630
1882
这个世界有上百万
08:56
around the world like Samantha, millions,
206
524512
2833
和萨曼莎一样的人，上百万，
08:59
and we've only begun to scratch the surface.
207
527345
3440
而我们其实才刚刚开始。
09:02
What we've done so far is we have
208
530785
1642
我们到目前为止所做的就是，
09:04
a few surrogate talkers from around the U.S.
209
532427
3859
我们有来自美国的几个代理说话者，
09:08
who have donated their voices,
210
536286
1507
他们捐献了自己的声音，
09:09
and we have been using those
211
537793
1928
而我们正在用这些声音
09:11
to build our first few personalized voices.
212
539721
4472
来打造最初的一些个性化的声音。
09:16
But there's so much more work to be done.
213
544193
1756
但是接下来的任务还很重。
09:17
For Samantha, her surrogate
214
545949
2188
就萨曼莎，她的代理说话者
09:20
came from somewhere in the Midwest, a stranger
215
548137
3046
来自中西部的一个地方，
09:23
who gave her the gift of voice.
216
551183
3841
一个将声音赠送给她的陌生人。
09:27
And as a scientist, I'm so excited
217
555024
2153
作为一名科学家，我很期待
09:29
to take this work out of the laboratory
218
557177
1935
将这项工作搬到实验室之外，
09:31
and finally into the real world
219
559112
1800
最终搬进现实世界
09:32
so it can have real-world impact.
220
560912
3165
并产生真正的影响。
09:36
What I want to share with you next
221
564077
1582
我接下来想跟你们分享的是
09:37
is how I envision taking this work
222
565659
2175
我对如何将这项工作
09:39
to that next level.
223
567834
2711
推进到下一个层次的展望。
09:42
I imagine a whole world of surrogate donors
224
570545
3887
我想象到一个充满了代理说话者的世界，
09:46
from all walks of life, different sizes, different ages,
225
574432
3260
他们来自不同的行业，有着不同的体型和年龄，
09:49
coming together in this voice drive
226
577692
3058
他们为这个声音计划走到一起，
09:52
to give people voices
227
580750
2270
希望赋予人们
09:55
that are as colorful as their personalities.
228
583020
3799
和他们的性格一样丰富多彩的声音。
09:58
To do that as a first step,
229
586819
2300
实现这个目标的第一步，
10:01
we've put together this website, VocaliD.org,
230
589119
3275
我们建立了一个网站：VocaliD.org，
10:04
as a way to bring together those
231
592394
1624
通过这个网站，我们把
10:06
who want to join us as voice donors,
232
594018
2675
愿意以声音捐献者或专业知识捐献者的身份
10:08
as expertise donors,
233
596693
1772
加入到我们的人们团结在一起，
10:10
in whatever way to make this vision a reality.
234
598465
5339
不管以何种方式，来一起实现这个愿景。
10:15
They say that giving blood can save lives.
235
603804
4153
人们说献血可以拯救生命。
10:19
Well, giving your voice can change lives.
236
607957
4982
那么，捐献您的声音可以改变生命。
10:24
All we need is a few hours of speech
237
612939
3050
我们需要的仅仅是几小时的
10:27
from our surrogate talker,
238
615989
1491
代理说话者的话语，
10:29
and as little as a vowel from our target talker,
239
617480
4733
以及目标说话者的一个小小的元音，
10:34
to create a unique vocal identity.
240
622213
3711
就可以打造一个独特的声音。
10:37
So that's the science behind what we're doing.
241
625924
2626
这就是我们所做的研究背后的科学。
10:40
I want to end by circling back to the human side
242
628550
4455
作为结尾，我还是想回到人的主题，
10:45
that is really the inspiration for this work.
243
633005
4102
这也是这项工作的真正灵感来源。
10:49
About five years ago, we built our very first voice
244
637107
3699
大约五年前，我们第一次给一个名为威廉的男孩
10:52
for a little boy named William.
245
640806
2501
打造了他的声音。
10:55
When his mom first heard this voice,
246
643307
2357
当他的妈妈第一次听到这个声音的时候，
10:57
she said, "This is what William
247
645664
2345
她说：“如果威廉
11:00
would have sounded like
248
648009
1546
可以讲话，
11:01
had he been able to speak."
249
649555
2449
他的声音就该是这样的。
11:04
And then I saw William typing a message
250
652004
2418
然后我看到威廉在他的设备上
11:06
on his device.
251
654422
1362
打出一条消息。
11:07
I wondered, what was he thinking?
252
655784
3293
我在想，他在想什么？
11:11
Imagine carrying around someone else's voice
253
659077
3590
想象一下九年来一直用着
11:14
for nine years
254
662667
2193
别人的声音，
11:16
and finally finding your own voice.
255
664860
4844
然后最终找到了你自己的声音。
11:21
Imagine that.
256
669704
1377
想象一下。
11:23
This is what William said:
257
671081
2797
威廉说的是：
11:25
"Never heard me before."
258
673878
4463
“我从来没有听过我自己的声音。”
11:32
Thank you.
259
680417
1619
谢谢。
11:34
(Applause)
260
682036
4724
（掌声）
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