How AI could compose a personalized soundtrack to your life | Pierre Barreau

167,508 views ・ 2018-10-01

TED


请双击下面的英文字幕来播放视频。

翻译人员: Ch'ng Tsu Pang 校对人员: Yolanda Zhang
00:12
About two and a half years ago, I watched this movie called "Her."
0
12913
3428
大概两年半前,我看了《她》这部电影。
00:16
And it features Samantha, a superintelligent form of AI
1
16365
5127
电影里的“女主角”沙曼塔 是极度聪明的人工智能,
00:21
that cannot take physical form.
2
21516
1760
但却并不拥有实体肉身。
00:23
And because she can't appear in photographs,
3
23858
2230
她没办法拥有自己的照片,
00:26
Samantha decides to write a piece of music
4
26112
2016
所以她决定写首曲子
00:28
that will capture a moment of her life just like a photograph would.
5
28152
3466
来捕捉她生命中的片段,正如拍照一般。
00:32
As a musician and an engineer, and someone raised in a family of artists,
6
32953
4408
我出生于艺术世家, 现在是一名音乐家兼工程师,
00:37
I thought that this idea of musical photographs was really powerful.
7
37385
4176
我觉得音乐摄影这个想法很强大。
00:41
And I decided to create an AI composer.
8
41955
2868
所以我决定制造一款会作曲的人工智能。
00:44
Her name is AIVA, and she's an artificial intelligence
9
44847
3488
这款人工智能的名字就叫做“AIVA”。
00:48
that has learned the art of music composition
10
48359
2393
通过阅读历史上最优秀的
00:50
by reading over 30,000 scores of history's greatest.
11
50776
2785
三万多首乐谱,她学会了作曲的艺术。
00:54
So here's what one score looks like to the algorithm
12
54165
2456
如果以这种矩阵似的形式来表现的话,
00:56
in a matrix-like representation.
13
56645
2202
在我们的算法看来, 乐谱就是长这样子的。
00:58
And here's what 30,000 scores,
14
58871
2246
而如果把莫扎特、贝多芬等人
01:01
written by the likes of Mozart and Beethoven,
15
61141
2096
写的三万首乐谱同置于一个框架之中,
01:03
look like in a single frame.
16
63261
2031
看起来就是这个样子了。
01:07
So, using deep neural networks, AIVA looks for patterns in the scores.
17
67609
4340
利用深层神经网络,AIVA 开始从这些乐谱中找出规律。
01:12
And from a couple of bars of existing music,
18
72260
3563
从现有音乐中的几个小节,
01:15
it actually tries to infer what notes should come next in those tracks.
19
75847
3708
她会推测乐曲中接下来该有什么音符。
01:19
And once AIVA gets good at those predictions,
20
79887
2498
一旦预测变得越来越准确,
01:22
it can actually build a set of mathematical rules
21
82409
3613
AIVA 就开始为该种曲风建立起
一套数学规则,
01:26
for that style of music
22
86046
1166
01:27
in order to create its own original compositions.
23
87236
2788
并以此写出她自己的原创歌曲。
01:30
And in a way, this is kind of how we, humans, compose music, too.
24
90490
3709
在某种程度上,这其实也是 我们人类作曲的方式。
01:34
It's a trial-and-error process,
25
94561
1492
这是一种反复试验的过程,
01:36
during which we may not get the right notes all the time.
26
96077
3008
我们未必每次都找得到最合适的音符,
01:39
But we can correct ourselves,
27
99109
1389
但我们可以用我们的
01:40
either with our musical ear or our musical knowledge.
28
100522
3293
音乐知识或耳朵来自我修正。
01:45
But for AIVA, this process is taken from years and years of learning,
29
105495
4143
这需要长时间的学习, 但 AIVA 把段漫长的学习时间,
01:49
decades of learning as an artist, as a musician and a composer,
30
109662
3262
也就是成为艺术家、音乐家、 作曲家所需的那几十年学习,
01:52
down to a couple of hours.
31
112948
1733
浓缩成只有几个小时。
01:55
But music is also a supersubjective art.
32
115030
2880
当然音乐也是个极为主观的艺术。
01:57
And we needed to teach AIVA
33
117934
1564
所以我们需要教导 AIVA
01:59
how to compose the right music for the right person,
34
119522
2445
如何为不同的人制作不同的音乐,
02:01
because people have different preferences.
35
121991
2141
毕竟每个人的喜好都有所不同。
02:04
And to do that, we show to the algorithm over 30 different category labels
36
124156
4190
为此,我们为数据库里的每一首乐谱 标上了三十余种不同的类别标签,
02:08
for each score in our database.
37
128370
1945
并展示给我们的算法。
02:10
So those category labels are like mood
38
130339
2516
这些类别标签包括音乐的气氛、
02:12
or note density or composer style of a piece
39
132879
2920
音符的密度、曲风
02:15
or the epoch during which it was written.
40
135823
2524
或作品的年代等等。
02:18
And by seeing all this data,
41
138371
1935
而通过这些数据,
02:20
AIVA can actually respond to very precise requirements.
42
140330
3056
AIVA 就可以满足一些很精确的要求。
02:23
Like the ones, for example, we had for a project recently,
43
143966
3103
比如,我们最近接了一个项目,
02:27
where we were commissioned to create a piece
44
147093
3035
要创作一首会让人想起
02:30
that would be reminiscent of a science-fiction film soundtrack.
45
150152
3027
科幻电影配乐的曲子。
02:33
And the piece that was created is called "Among the Stars"
46
153839
4500
于是我们就作了一首曲子, 叫《众星之中》,
02:38
and it was recorded with CMG Orchestra in Hollywood,
47
158363
3317
后来由指挥家约翰 · 比尔领衔的 好莱坞 CMG 管弦乐队
02:41
under great conductor John Beal,
48
161704
1714
录制了这首曲子。
02:43
and this is what they recorded, made by AIVA.
49
163442
3067
这就是他们录制好的,AIVA 的作品。
02:47
(Music)
50
167657
7000
(音乐)
03:30
(Music ends)
51
210196
2190
(音乐结束)
03:34
What do you think?
52
214719
1207
你们觉得怎样?
03:35
(Applause)
53
215950
4380
(掌声)
03:40
Thank you.
54
220354
1150
谢谢。
03:42
So, as you've seen, AI can create beautiful pieces of music,
55
222001
4055
你们都看到了,人工智能 可以写出漂亮的乐章,
03:46
and the best part of it
56
226080
1803
而最棒的是,
03:47
is that humans can actually bring them to life.
57
227907
2581
人类可以把这音乐演奏出来。
03:51
And it's not the first time in history
58
231506
1838
科技可以扩展人类的创造力,
03:53
that technology has augmented human creativity.
59
233368
3344
这其实并不新鲜。
03:56
Live music was almost always used in silent films
60
236736
3072
以前的默片就经常会有现场伴奏,
03:59
to augment the experience.
61
239832
1666
好带给观众更好的观影经验。
04:01
But the problem with live music is that it didn't scale.
62
241522
3182
但现场伴奏有个问题: 它没办法适应不同场景的需要。
04:04
It's really hard to cram a full symphony into a small theater,
63
244728
3429
要把整个交响乐团塞进一个 小剧院里面已经很不容易,
04:08
and it's really hard to do that for every theater in the world.
64
248181
2991
要让全世界所有的剧院 都有交响乐团,那就更难了。
04:11
So when music recording was actually invented,
65
251196
2849
所以当人类发明了录音,
04:14
it allowed content creators, like film creators,
66
254069
2349
所有的内容创作者,比如电影创作者,
04:16
to have prerecorded and original music
67
256442
2786
便可以为他们故事里的每一个画面
04:19
tailored to each and every frame of their stories.
68
259252
2873
量身定制原创音乐,并且预先录制好。
04:22
And that was really an enhancer of creativity.
69
262149
2626
这无疑强化了我们的创造力。
04:26
Two and a half years ago, when I watched this movie "Her,"
70
266617
3286
两年半前,我在看《她》这部电影的时候,
04:29
I thought to myself that personalized music
71
269927
2907
不禁心想到,个人化的音乐
04:32
would be the next single biggest change in how we consume and create music.
72
272858
4417
将会是我们未来创作和消费 音乐时最重大的一个改变。
04:38
Because nowadays, we have interactive content, like video games,
73
278187
4222
因为如今,我们有许多 互动性的内容,比如说电子游戏,
04:42
that have hundreds of hours of interactive game plays,
74
282433
3015
里面有上百个小时的互动式游戏内容,
04:45
but only two hours of music, on average.
75
285472
2113
平均却只配有两个小时的音乐。
04:47
And it means that the music loops and loops and loops
76
287609
2492
也就是说,游戏的音乐 是不停地循环重播,
04:50
over and over again, and it's not very immersive.
77
290125
2332
而且没法让人真正沉浸其中。
04:52
So what we're working on is to make sure that AI can compose
78
292467
3951
我们目前正在努力让人工智能创作出
04:56
hundreds of hours of personalized music
79
296442
2302
上百个小时的个性化音乐,
04:58
for those use cases where human creativity doesn't scale.
80
298768
3866
以填补那些人类创造力无法涵盖的领域。
05:03
And we don't just want to do that for games.
81
303363
2206
而且我们想做的绝不限于电子游戏。
05:06
Beethoven actually wrote a piece for his beloved, called "Für Elise,"
82
306657
4428
贝多芬曾经给他的挚爱 写过一首《致爱丽丝》,
05:11
and imagine if we could bring back Beethoven to life.
83
311109
3651
想像一下,如果我们可以让贝多芬重生,
05:14
And if he was sitting next to you, composing a music for your personality
84
314784
5230
让他坐在你身旁,依据 你的性情和人生故事,
专门谱写一首曲子。
05:20
and your life story.
85
320038
1333
05:22
Or imagine if someone like Martin Luther King, for example,
86
322632
2779
或者你也可以想象一下某个人, 比如说马丁 · 路德 · 金,
05:25
had a personalized AI composer.
87
325435
2039
如果他拥有一款个人化的、 会作曲的人工智能。
05:27
Maybe then we would remember
88
327498
1373
那《我有一个梦想》对我们而言,
05:28
"I Have a Dream" not only as a great speech,
89
328895
2056
除了是一次伟大的演讲,
05:30
but also as a great piece of music, part of our history,
90
330975
2626
或许也会是我们历史上一首可以体现
05:33
and capturing Dr. King's ideals.
91
333625
2133
金博士理想的伟大音乐作品。
05:36
And this is our vision at AIVA:
92
336069
1889
我们对 AIVA 的期许就是:
05:37
to personalize music so that each and every one of you
93
337982
2578
让她为你们在座的每一位, 甚至世界上的每一个人
05:40
and every individual in the world
94
340584
1649
量身创作音乐。
05:42
can have access to a personalized live soundtrack,
95
342257
3074
让每个人可以根据 自己的人生故事和性格特点,
05:45
based on their story and their personality.
96
345355
2505
找到专属于自己的现场配乐。
05:49
So this moment here together at TED is now part of our life story.
97
349915
4079
现在我们聚集在 TED 会场, 这也是我们人生经历的一部分,
05:54
So it only felt fitting that AIVA would compose music for this moment.
98
354018
4169
如果 AIVA 可以为此谱曲一首的话, 那就再适合不过了。
05:58
And that's exactly what we did.
99
358674
2444
我们还真的就这么做了。
06:01
So my team and I worked on biasing AIVA on the style of the TED jingle,
100
361436
5022
我和团队让 AIVA 模拟 TED 的广告曲风,
06:06
and on music that makes us feel a sense of awe and wonder.
101
366482
3444
以及那些会让我们惊叹 或心生敬畏的音乐风格。
06:09
And the result is called "The Age of Amazement."
102
369950
3913
于是我们就有了这首《惊奇的年代》。
06:13
Didn't take an AI to figure that one out.
103
373887
2396
当然这曲名不是人工智能想出来的。
06:16
(Laughter)
104
376307
1150
(笑声)
06:18
And I couldn't be more proud to show it to you,
105
378152
2402
今天可以给大家带来这首曲子, 我感到非常自豪,
06:20
so if you can, close your eyes and enjoy the music.
106
380578
2428
可以的话,请闭上双眼, 好好享受一下这音乐吧。
06:23
Thank you very much.
107
383030
1333
谢谢。
06:25
(Music)
108
385537
4195
(音乐)
06:35
[The Age of Amazement Composed by AIVA]
109
395176
2667
【《惊奇的年代》 作曲:AIVA】
08:19
(Music ends)
110
499649
1151
(音乐结束)
08:20
This was for all of you.
111
500824
1373
谨以此曲献给大家。
08:22
Thank you.
112
502221
1166
谢谢。
08:23
(Applause)
113
503411
4573
(掌声)
关于本网站

这个网站将向你介绍对学习英语有用的YouTube视频。你将看到来自世界各地的一流教师教授的英语课程。双击每个视频页面上显示的英文字幕,即可从那里播放视频。字幕会随着视频的播放而同步滚动。如果你有任何意见或要求,请使用此联系表与我们联系。

https://forms.gle/WvT1wiN1qDtmnspy7