Karen Lellouche Tordjman: Siri, Alexa, Google ... what comes next? | TED

59,601 views ใƒป 2022-02-21

TED


ืื ื ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ืœืžื˜ื” ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ.

ืชืจื’ื•ื: Jenny Rublevsky ืขืจื™ื›ื”: zeeva livshitz
00:04
It's Monday morning.
0
4334
1376
ืขื›ืฉื™ื• ืฉื ื™ ื‘ื‘ื•ืงืจ.
00:05
You've just received a voice notification from your voice assistant.
1
5752
4087
ื”ืจื’ืข ืงื™ื‘ืœืชื ื”ื•ื“ืขื” ืงื•ืœื™ืช ืžื”ืขื•ื–ืจืช ื”ืงื•ืœื™ืช ืฉืœื›ื.
00:11
Voice assistant: Good morning.
2
11132
1460
ืขื•ื–ืจืช ืงื•ืœื™ืช: ื‘ื•ืงืจ ื˜ื•ื‘.
00:12
There's a new pair of pants on sale at your favorite store.
3
12592
2836
ื™ืฉ ื–ื•ื’ ืžื›ื ืกื™ื™ื ื—ื“ืฉ ื‘ืžื‘ืฆืข ื‘ื—ื ื•ืช ื”ืื”ื•ื‘ื” ืขืœื™ื›ื.
00:15
Last week, while wearing your smart glasses,
4
15428
2169
ื‘ืฉื‘ื•ืข ืฉืขื‘ืจ, ื‘ื–ืžืŸ ืฉื”ืจื›ื‘ืชื ืืช ื”ืžืฉืงืคื™ื™ื ื”ื—ื›ืžื•ืช,
00:17
you were drawn to a similar pair on a colleague.
5
17597
2461
ื ืžืฉื›ืชื ืœื–ื•ื’ ื“ื•ืžื” ืืฆืœ ืงื•ืœื’ื”.
00:20
Would you like to wear them to the party on Thursday?
6
20100
2544
ื”ืื ืชืจืฆื• ืœืœื‘ื•ืฉ ืื•ืชื ืœืžืกื™ื‘ื” ื‘ื™ื•ื ื—ืžื™ืฉื™?
00:23
I suggest you skip the movies tomorrow night to afford them.
7
23353
3170
ืื ื™ ืžืฆื™ืขื” ืœื›ื ืœื“ืœื’ ืขืœ ื”ืกืจื˜ ืžื—ืจ ื‘ืขืจื‘ ื›ื“ื™ ืฉืชื•ื›ืœื• ืœื”ืจืฉื•ืช ืื•ืชื ืœืขืฆืžื›ื.
00:26
Say "yes" to validate the purchase.
8
26565
2335
ืืžืจื• โ€œื›ืŸโ€ ื›ื“ื™ ืœืืฉืจ ืืช ื”ืจื›ื™ืฉื”.
00:29
Let me know if I should also find a gift for your host.
9
29484
3087
ืชื™ื™ื“ืขื• ืื•ืชื™ ืื ืœืžืฆื•ื ืžืชื ื” ืขื‘ื•ืจ ื”ืžืืจื— ืฉืœื›ื.
00:32
I can make suggestions based on your budget
10
32612
2461
ืื ื™ ื™ื›ื•ืœื” ืœื”ื›ื™ืŸ ื”ืฆืขื•ืช ื”ืžื‘ื•ืกืกื•ืช ืขืœ ื”ืชืงืฆื™ื‘
00:35
and her public Instagram profile.
11
35115
2461
ื•ืขืœ ื—ืฉื‘ื•ืŸ ื”ืื™ื ืกื˜ื’ืจื ืฉืœืš.
00:37
Say "yes" to proceed.
12
37576
1835
ืืžืจื• โ€œื›ืŸโ€ ื›ื“ื™ ืœื”ืžืฉื™ืš.
00:40
Karen Lellouche Tordjman: Yes.
13
40161
1460
ืงืจืŸ ืœืœื•ืฉ ืชื•ืจื’โ€™ืžืŸ: ื›ืŸ.
00:41
So, thrilling or scary?
14
41913
4463
ืื–, ืžืจื’ืฉ ืื• ืžืคื—ื™ื“?
00:46
Because this is how our future could look very soon.
15
46793
3629
ื‘ื’ืœืœ ืฉื›ืš ื”ืขืชื™ื“ ืฉืœื ื• ื™ื›ื•ืœ ืœื”ื™ืจืื•ืช ื‘ืงืจื•ื‘ ืžืื•ื“.
00:51
As a working mom, I find it exciting.
16
51089
2836
ื›ืืžื ืขื•ื‘ื“ืช, ืื ื™ ืžืชืจื’ืฉืช ืžื›ืš.
00:54
I would love to stop doing Google searches,
17
54593
3503
ืื ื™ ืืฉืžื— ืœื”ืคืกื™ืง ืœืขืฉื•ืช ื—ื™ืคื•ืฉื™ื ื‘ื’ื•ื’ืœ,
00:58
wasting time doing Amazon scrolling, budget calculation
18
58138
4337
ืœื‘ื–ื‘ื– ื–ืžืŸ ื‘ื’ืœื™ืœื” ื‘ืืžื–ื•ืŸ, ืœื—ืฉื‘ ืชืงืฆื™ื‘
01:02
or optimizing my calendar.
19
62517
2628
ืื• ืœื™ื™ืขืœ ืืช ืœื•ื— ื”ืฉื ื” ืฉืœื™.
01:05
What is thrilling is the prospect of having a companion
20
65687
4504
ืžื” ืฉืžืจื’ืฉ ื”ื•ื ื”ืจืขื™ื•ืŸ ืฉื™ืฉ ืœื™ ืฉื•ืชืฃ
01:10
that would cater specifically to my needs and requests.
21
70191
3712
ืฉื™ื•ื›ืœ ืœืกืคืง ืกืคืฆื™ืคื™ืช ืืช ื”ืฆืจื›ื™ื ื•ื”ื‘ืงืฉื•ืช ืฉืœื™.
01:13
Just imagine, it could do things, like using my heart rate
22
73945
4713
ืจืง ืชื“ืžื™ื™ื ื•, ื–ื” ื™ื•ื›ืœ ืœืขืฉื•ืช ื“ื‘ืจื™ื, ื›ืžื• ืœื”ืฉืชืžืฉ ื‘ืงืฆื‘ ื”ืœื‘ ืฉืœื™
01:18
to tweak my Starbucks order to reduce caffeine.
23
78700
3295
ื›ื“ื™ ืœื›ื•ื•ื ืŸ ืืช ื”ื–ืžื ืช ื”ืกื˜ืจื‘ืืงืก ืฉืœื™ ืขืœ ืžื ืช ืœื”ืคื—ื™ืช ื‘ืงืคืื™ืŸ.
01:22
It could take into account my lunch and the number of steps I've walked
24
82662
5047
ื–ื” ื™ื•ื›ืœ ืœืงื—ืช ื‘ื—ืฉื‘ื•ืŸ ืืช ืืจื•ื—ืช ื”ืฆื”ืจื™ื™ื ืฉืœื™ ื•ืืช ืžืกืคืจ ื”ืฆืขื“ื™ื ืฉืœื™
01:27
to tailor a workout for me.
25
87709
2377
ื›ื“ื™ ืœืชื›ื ืŸ ืื™ืžื•ืŸ ื‘ืฉื‘ื™ืœื™.
01:30
It could even align with my friend's smart assistant
26
90920
3796
ื–ื” ืืคื™ืœื• ื™ื•ื›ืœ ืœื”ืกืชื ื›ืจืŸ ืขื ื”ืขื•ื–ืจ ื”ื—ื›ื ืฉืœ ื—ื‘ืจื™ื™
01:34
to craft evening plans
27
94716
1668
ื›ื“ื™ ืœืชื›ื ืŸ ืชื›ื ื™ื•ืช ืœืขืจื‘
01:36
that would fit everyone's budget, calendars and locations.
28
96426
4087
ืฉื™ื›ื•ืœื•ืช ืœื”ืชืื™ื ืœืชืงืฆื™ื‘, ืœืœื•ื— ื”ืฉื ื” ื•ืœืžื™ืงื•ื ืฉืœ ื›ื•ืœื,
01:42
Obviously, this scenario could easily lean towards the scary.
29
102140
3795
ื›ืžื•ื‘ืŸ, ื”ืชืจื—ื™ืฉ ื”ื–ื” ื™ื›ื•ืœ ืœื”ื™ื•ืช ื’ื ืžืคื—ื™ื“.
01:46
This is why there needs to be regulation in place.
30
106519
2920
ืœื›ืŸ ืฆืจื™ื›ื” ืœื”ื™ื•ืช ืจื’ื•ืœืฆื™ื” ื‘ื ื•ืฉื.
01:49
All users, all consumers, should always remain in full control of data they share
31
109898
6631
ื›ืœ ื”ืžืฉืชืžืฉื™ื, ื›ืœ ื”ืฆืจื›ื ื™ื, ื™ื™ืฉืืจื• ืชืžื™ื“ ื‘ืฉืœื™ื˜ื” ืžื•ื—ืœื˜ืช ืขืœ ื”ืžื™ื“ืข ืฉื”ื ืžืฉืชืคื™ื
01:56
and on which type of recommendations they agree to get or not to get.
32
116571
5089
ื•ืขืœ ืื™ื–ื” ืกื•ื’ ืฉืœ ื”ืžืœืฆื•ืช ื”ื ืžืกื›ื™ืžื™ื ืœืงื‘ืœ ืื• ืœื ืœืงื‘ืœ.
02:03
Now, recommendation engines powered by artificial intelligence are not new.
33
123286
5506
ืžื ื•ืขื™ ื”ืžืœืฆื•ืช ื”ืžื•ื ืขื™ื ืขืœ ื™ื“ื™ ื‘ื™ื ื” ืžืœืื›ื•ืชื™ืช ืื™ื ื ื—ื“ืฉื™ื.
02:09
Far from it.
34
129167
1418
ื”ืจื—ืง ืžื›ืš.
02:10
We actually use them multiple times a day already.
35
130627
3420
ืœืžืขืฉื” ืื ื—ื ื• ืžืฉืชืžืฉื™ื ื‘ื”ื ืคืขืžื™ื ืจื‘ื•ืช ื‘ื™ื•ื ื›ื‘ืจ ืขื›ืฉื™ื•.
02:14
On average, 70 percent of the time spent on YouTube
36
134756
4588
ื‘ืžืžื•ืฆืข, 70% ืžื”ื–ืžืŸ ื‘ื™ื•ื˜ื™ื•ื‘
02:19
is on videos recommended by their algorithm.
37
139344
3003
ื”ื•ื ื‘ืกืจื˜ื•ื ื™ื ื”ืžื•ืžืœืฆื™ื ืขืœ ื™ื“ื™ ืืœื’ื•ืจื™ืชื.
02:23
And you may also own a wristband that you use to track your sleep
38
143139
4213
ื•ืื•ืœื™ ื™ืฉ ืœื›ื ืฆืžื™ื“ ืฉืขื•ืงื‘ ืื—ืจื™ ื”ืฉื™ื ื” ืฉืœื›ื
02:27
or monitor your workout.
39
147394
2752
ืื• ืžืคืงื— ืขืœ ื”ืื™ืžื•ืŸ ืฉืœื›ื.
02:31
And there's the first generation of voice-enabled virtual assistants.
40
151189
4379
ื•ื™ืฉ ืืช ื”ื“ื•ืจ ื”ืจืืฉื•ืŸ ืฉืœ ื”ืขื•ื–ืจื•ืช ื”ื•ื™ืจื˜ื•ืืœื™ื•ืช ื”ืžื•ืคืขืœื•ืช ื‘ืงื•ืœ.
02:35
You know, like Google Home or Amazon Alexa
41
155610
3420
ืืชื ื™ื•ื“ืขื™ื, ื›ืžื• ื’ื•ื’ืœ ื”ื•ื ืื• ืืžื–ื•ืŸ ืืœื›ืกื”
02:39
that you can use to change the temperature of your room.
42
159072
2836
ืฉืืชื ื™ื›ื•ืœื™ื ืœื”ืฉืชืžืฉ ื‘ื”ื ื›ื“ื™ ืœืฉื ื•ืช ืืช ื”ื˜ืžืคืจื˜ื•ืจื” ื‘ื—ื“ืจ.
02:42
Your car as well is probably equipped with the same things
43
162450
6548
ื”ืžื›ื•ื ื™ืช ืฉืœื›ื ื’ื ื›ื ืจืื” ื›ื‘ืจ ืžืื•ื‘ื–ืจืช ื‘ืื•ืชื ื”ื“ื‘ืจื™ื
02:49
that can help you manage your music
44
169040
2211
ืฉื™ื›ื•ืœื™ื ืœืขื–ื•ืจ ืœื›ื ืœื ื”ืœ ืืช ื”ืžื•ื–ื™ืงื” ืฉืœื›ื
02:51
or give you directions, hands-free.
45
171292
2586
ืื• ืœื”ื›ื•ื•ื™ืŸ ืืชื›ื, ืœืœื ืžื’ืข.
02:55
But there is one thing that all these tools have in common.
46
175839
3420
ืื‘ืœ ื™ืฉ ื“ื‘ืจ ืื—ื“ ืฉื™ืฉ ืœื›ืœ ืืœื” ื‘ืžืฉื•ืชืฃ.
02:59
They leverage AI to help you in one specific area of your life.
47
179718
6256
ื”ื ืžืฉืชืžืฉื™ื ื‘ื‘ื™ื ื” ืžืœืื›ื•ืชื™ืช ืฉืชืขื–ื•ืจ ืœื›ื ื‘ืชื—ื•ื ืžืกื•ื™ื ื‘ื—ื™ื™ื.
03:06
Your home, your car, your health.
48
186015
4255
ื”ื‘ื™ืช ืฉืœื›ื, ื”ืžื›ื•ื ื™ืช ืฉืœื›ื, ื”ื‘ืจื™ืื•ืช ืฉืœื›ื.
03:10
They stay in the lane.
49
190645
2044
ื”ื ื ืฉืืจื™ื ื‘ืžืกืœื•ืœ.
03:13
Now imagine a new generation of voice assistants
50
193440
5005
ืขื›ืฉื™ื• ื“ืžื™ื™ื ื• ื“ื•ืจ ื—ื“ืฉ ืฉืœ ืขื•ื–ืจื•ืช ืงื•ืœื™ื•ืช
03:18
that crosses all lanes,
51
198486
2670
ืฉืขื•ืœื•ืช ืขืœ ื›ืœ ื”ืžืกืœื•ืœื™ื,
03:21
that synchronizes everything.
52
201156
2544
ืฉืžืกื ื›ืจื ื•ืช ืืช ื”ื›ืœ.
03:24
So you may ask, why hasn't this happened already?
53
204200
3462
ืื– ืื•ืœื™ ืชืฉืืœื•, ืœืžื” ื–ื” ืœื ืงืจื” ืขื“ื™ื™ืŸ?
03:28
Because there are two technological bricks that are critical to make this happen
54
208413
4463
ื‘ื’ืœืœ ืฉืฉืชื™ ืœื‘ื ื™ื ื˜ื›ื ื•ืœื•ื’ื™ื•ืช ืฉืงืจื™ื˜ื™ื•ืช ื›ื“ื™ ืœื’ืจื•ื ืœื–ื” ืœืงืจื•ืช
03:32
that are still missing.
55
212917
1502
ืขื“ื™ื™ืŸ ื—ืกืจื•ืช.
03:34
One is voice.
56
214794
2252
ืื—ืช ื”ื™ื ืงื•ืœ.
03:37
These tools must be able to understand everything we say.
57
217589
4045
ื”ื›ืœื™ื ื”ืืœื” ื—ื™ื™ื‘ื™ื ืœื”ื‘ื™ืŸ ืืช ื›ืœ ืžื” ืฉืื ื—ื ื• ืื•ืžืจื™ื.
03:41
And clearly, this is not the case today.
58
221676
2336
ื•ื›ืžื•ื‘ืŸ, ื–ื” ืœื ื”ืžืงืจื” ื›ื™ื•ื.
03:44
For my kids, maybe yours as well,
59
224387
2586
ื‘ืฉื‘ื™ืœ ื”ื™ืœื“ื™ื ืฉืœื™, ื•ืื•ืœื™ ื’ื ืฉืœื›ื,
03:47
Siri is still an endless source of fun.
60
227015
3503
ืกื™ืจื™ ื”ื™ื ืขื“ื™ื™ืŸ ืžืงื•ืจ ืื™ื ืกื•ืคื™ ืœื›ื™ืฃ.
03:51
They like asking very simple questions
61
231269
2669
ื”ื ืื•ื”ื‘ื™ื ืœืฉืื•ืœ ืฉืืœื•ืช ืžืื•ื“ ืคืฉื•ื˜ื•ืช
03:53
and still get very confused answers.
62
233980
3337
ื•ืขื“ื™ื™ืŸ ืžืงื‘ืœื™ื ืชืฉื•ื‘ื•ืช ืžื‘ื•ืœื‘ืœื•ืช ืžืื•ื“.
03:59
Two, breadth.
63
239194
2252
ืฉืชื™ื™ื, ืจื•ื—ื‘.
04:02
They must be able to provide a large range of recommendations
64
242530
3879
ื”ื ื—ื™ื™ื‘ื™ื ืœื”ื™ื•ืช ืžืกื•ื’ืœื™ื ืœืกืคืง ื˜ื•ื•ื— ืจื—ื‘ ืฉืœ ื”ืžืœืฆื•ืช
04:06
to cover whatever we may need.
65
246451
2461
ื›ื“ื™ ืœื›ืกื•ืช ืืช ื›ืœ ืžื” ืฉื ืฆื˜ืจืš.
04:09
So tech players are still working on cracking those two elements,
66
249746
4421
ืื– ืฉื—ืงื ื™ื ื˜ื›ื ื™ื™ื ืขื“ื™ื™ืŸ ืขื•ื‘ื“ื™ื ื›ื“ื™ ืœืคืฆื— ืืช ืฉื ื™ ื”ืืœืžื ื˜ื™ื ื”ืืœื”
04:14
voice and breadth.
67
254167
1793
ืงื•ืœ ื•ืจื•ื—ื‘.
04:17
On voice, as I said, Alexa, Google Home, Siri and the likes,
68
257045
6089
ืœื’ื‘ื™ ืงื•ืœ, ื›ืžื• ืฉืืžืจืชื™, ืืœื›ืกื”, ื’ื•ื’ืœ ื”ื•ื, ืกื™ืจื™ ื•ื“ื•ืžื™ื”ื,
04:23
they don't understand us entirely and systematically for now.
69
263176
4838
ื”ื ืœื ืžื‘ื™ื ื™ื ืื•ืชื ื• ืœื’ืžืจื™ ื•ื‘ืื•ืคืŸ ืฉื™ื˜ืชื™ ื ื›ื•ืŸ ืœืขื›ืฉื™ื•.
04:28
Actually, understanding human language is difficult.
70
268723
3253
ืœืžืขืฉื”, ืœื”ื‘ื™ืŸ ืืช ื”ืฉืคื” ื”ืื ื•ืฉื™ืช ื–ื” ืงืฉื”.
04:32
It's as much about the context
71
272477
2085
ื–ื” ืงืฉื•ืจ ืœื”ืงืฉืจ
04:34
as it is about the words themselves.
72
274604
2461
ื›ืžื• ืฉื–ื” ืงืฉื•ืจ ืœืžื™ืœื™ื ืขืฆืžืŸ.
04:37
And think about the accents or background noise.
73
277524
3545
ื•ืชื—ืฉื‘ื• ืขืœ ื”ืžื‘ื˜ื ืื• ืขืœ ืจืขืฉื™ ื”ืจืงืข.
04:41
It's already very difficult for Americans
74
281444
2586
ื–ื” ื›ื‘ืจ ืžืื•ื“ ืงืฉื” ืœืืžืจื™ืงืื™ื
04:44
to understand French people speaking English,
75
284030
4171
ืœื”ื‘ื™ืŸ ืฆืจืคืชื™ื ืฉืžื“ื‘ืจื™ื ืื ื’ืœื™ืช,
04:48
(Laughter)
76
288243
1042
(ืฆื—ื•ืง)
04:49
so can you imagine how difficult it can be for a robot?
77
289327
3420
ืื– ืืชื ื™ื›ื•ืœื™ื ืœื“ืžื™ื™ืŸ ื›ืžื” ืงืฉื” ื–ื” ื™ื›ื•ืœ ืœื”ื™ื•ืช ืœืจื•ื‘ื•ื˜?
04:53
So tech players are working hard on this.
78
293289
3170
ืื– ืฉื—ืงื ื™ื ื˜ื›ื ื™ื™ื ืขื•ื‘ื“ื™ื ืงืฉื” ืขืœ ื–ื”.
04:56
In 2018, Google launched an investment program
79
296918
4922
ื‘ืฉื ืช 2018, ื’ื•ื’ืœ ื”ืฉื™ืงื” ืชื•ื›ื ื™ืช ื”ืฉืงืขื”
05:01
for start-ups that would work with their Google Home suites.
80
301840
4879
ืœืกื˜ืืจื˜ ืืคื™ื ืฉืชืขื‘ื•ื“ ืขื ืžืขืจื›ื•ืช ื”ื’ื•ื’ืœ ื”ื•ื ืฉืœื”ื.
05:07
And since then, they have invested in over 15 companies.
81
307136
3504
ื•ืžืื–, ื”ื ื”ืฉืงื™ืขื• ื‘ื™ื•ืชืจ ืž-15 ื—ื‘ืจื•ืช.
05:11
Amazon has 10,000 employees
82
311307
4421
ืืžื–ื•ืŸ ืžืขืกื™ืงื” 10,000 ืขื•ื‘ื“ื™ื
05:15
working on Alexa voice technologies.
83
315728
2586
ืฉืขื•ื‘ื“ื™ื ืขืœ ื˜ื›ื ื•ืœื•ื’ื™ื•ืช ื”ืงื•ืœ ืฉืœ ืืœื›ืกื”.
05:18
So they will eventually crack the voice issues.
84
318982
3128
ื›ื“ื™ ืฉืœื‘ืกื•ืฃ ื”ื ื™ืฆืœื™ื—ื• ืœืคืฆื— ืืช ื‘ืขื™ื•ืช ื”ืงื•ืœ.
05:22
The smart assistants will be able to understand what we say,
85
322610
4088
ื”ืขื•ื–ืจื•ืช ื”ืงื•ืœื™ื•ืช ื™ื•ื›ืœื• ืœื”ื‘ื™ืŸ ืžื” ืื ื—ื ื• ืื•ืžืจื™ื,
05:26
the meaning of the words in their context.
86
326739
3087
ืืช ืžืฉืžืขื•ืช ื”ืžื™ืœื™ื ื‘ื”ืงืฉืจืŸ.
05:30
For example, if Iโ€™m using my smart assistant, and Iโ€™m listening to music,
87
330326
6340
ืœื“ื•ื’ืžื”, ืื ืื ื™ ืžืฉืชืžืฉืช ื‘ืขื•ื–ืจืช ื”ื—ื›ืžื” ืฉืœื™, ื•ืื ื™ ืžืื–ื™ื ื” ืœืžื•ื–ื™ืงื”,
05:36
and then I say, "Change."
88
336708
2294
ื•ืื– ืื ื™ ืื•ืžืจืช, โ€œืชืฉื ื™โ€œ.
05:39
VA: Hi, Karen.
89
339586
1126
ืขื•ื–ืจืช ืงื•ืœื™ืช: ื”ื™, ืงืจืŸ.
05:40
Do you want to change the temperature of the room?
90
340753
2795
ื”ืื ืืช ืจื•ืฆื” ืœืฉื ื•ืช ืืช ื”ื˜ืžืคืจื˜ื•ืจื” ื‘ื—ื“ืจ?
05:43
KLT: Well, clearly this is not working yet,
91
343548
2502
ืงืจืŸ: ื•ื‘ื›ืŸ, ื–ื” ื‘ื”ื—ืœื˜ ืœื ืขื•ื‘ื“ ืขื“ื™ื™ืŸ,
05:46
but in the future,
92
346092
1376
ืืš ื‘ืขืชื™ื“,
05:47
it will understand that I'm talking about changing the track,
93
347510
3295
ื–ื” ื™ื‘ื™ืŸ ืฉืื ื™ ืžื“ื‘ืจืช ืขืœ ืœื”ื—ืœื™ืฃ ืืช ื”ืฉื™ืจ,
05:50
not changing the temperature of the room.
94
350805
2169
ื•ืœื ืœืฉื ื•ืช ืืช ื”ื˜ืžืคืจื˜ื•ืจื” ื‘ื—ื“ืจ.
05:54
It will also understand our long and complex requests.
95
354100
4171
ื–ื” ื’ื ื™ื‘ื™ืŸ ืืช ื”ื‘ืงืฉื•ืช ื”ืืจื•ื›ื•ืช ื•ื”ืžืกื•ื‘ื›ื•ืช ืฉืœื ื•.
05:58
You know, when we start saying something
96
358313
2294
ืืชื ื™ื•ื“ืขื™ื, ื›ืฉืื ื—ื ื• ืžืชื—ื™ืœื™ื ืœื”ื’ื™ื“ ืžืฉื”ื•
06:00
and then we change our mind mid-sentence.
97
360607
2752
ื•ืื– ืื ื—ื ื• ืžืฉื ื™ื ืืช ื“ืขืชื ื• ื‘ืืžืฆืข ื”ืžืฉืคื˜.
06:04
But that doesn't stop here.
98
364861
1877
ืืš ื–ื” ืœื ืžืคืกื™ืง ื›ืืŸ.
06:06
With far-field speech recognition,
99
366779
3420
ืขื ื–ื™ื”ื•ื™ ืงื•ืœื™ ืจื—ื‘ ื˜ื•ื•ื—,
06:10
you will be able to use it from a distance,
100
370241
2544
ืืชื ืชื•ื›ืœื• ืœื”ืฉืชืžืฉ ื‘ื• ืžืžืจื—ืง,
06:12
from a room to another.
101
372827
1877
ืžื—ื“ืจ ืื—ื“ ืœืื—ืจ.
06:14
Even with background noise like kids screaming or traffic.
102
374746
4045
ืืคื™ืœื• ืขื ืจืขืฉ ืจืงืข ื›ืžื• ื™ืœื“ื™ื ืฆื•ืขืงื™ื ืื• ืชื ื•ืขื”.
06:19
(Kids shout, whistle blows)
103
379292
5756
(ื™ืœื“ื™ื ืฆื•ืขืงื™ื, ืžืฉืจื•ืงื™ืช ืฉื•ืจืงืช)
06:25
Tremendous progress has been made on this recently,
104
385673
3462
ื”ืชืงื“ืžื•ืช ืขืฆื•ืžื” ื”ืชืจื—ืฉื” ืœืื—ืจื•ื ื”,
06:29
largely due to Amazon's efforts on their Echo speaker technology.
105
389135
5380
ื‘ืขื™ืงืจ ืขืงื‘ ื”ืžืืžืฆื™ื ืฉืœ ืืžื–ื•ืŸ ืขืœ ื˜ื›ื ื•ืœื•ื’ื™ื™ืช ืจืžืงื•ืœ ื”-โ€œืืงื•โ€ ืฉืœื”ื.
06:36
Not only that.
106
396100
1627
ืœื ืจืง ื–ื”.
06:37
It will be able to understand in which mood youโ€™re in --
107
397727
3670
ื–ื” ื™ื”ื™ื” ืžืกื•ื’ืœ ืœื”ื‘ื™ืŸ ื‘ืื™ื–ื” ืžืฆื‘ ืจื•ื— ืืชื --
06:41
joy, sadness, annoyance --
108
401981
3295
ืฉืžื—ื”, ืขืฆื‘, ืจื•ื’ื– --
06:45
and will be able to mimic these feelings too.
109
405276
3337
ื•ื™ื”ื™ื” ืžืกื•ื’ืœ ืœื—ืงื•ืช ืืช ื”ืจื’ืฉื•ืช ื”ืืœื” ื’ื.
06:49
And as natural language processing advances further --
110
409989
4630
ื•ื›ื›ืœ ืฉืขื™ื‘ื•ื“ ืฉืคื” ื˜ื‘ืขื™ ืžืชืงื“ื --
06:54
so natural language processing is the technology behind this --
111
414661
3461
ื›ืš ื’ื ืขื™ื‘ื•ื“ ืฉืคื” ื˜ื‘ืขื™ ื”ื•ื ื”ื˜ื›ื ื•ืœื•ื’ื™ื” ืžืื—ื•ืจื™ ื–ื” --
06:58
so as it advances further,
112
418164
3253
ื›ืš ืฉื›ื›ืœ ืฉื–ื” ืžืชืงื“ื,
07:01
the voice-enabled interactions will increasingly be refined.
113
421459
4546
ื”ืื™ื ื˜ืจืืงืฆื™ื•ืช ื”ืžื•ืคืขืœื•ืช ื‘ืงื•ืœ ื™ืžืฉื™ื›ื• ืœื”ืชืคืชื—.
07:07
Now, the second challenge and the biggest, in my opinion,
114
427215
4671
ื›ืขืช, ื”ืืชื’ืจ ื”ืฉื ื™ ื•ื”ื’ื“ื•ืœ, ืœื“ืขืชื™,
07:11
revolves around the breadth of recommendations provided.
115
431928
4713
ืกื•ื‘ื‘ ืกื‘ื™ื‘ ื”ืจื•ื—ื‘ ืฉืœ ื”ื”ืžืœืฆื•ืช ื”ื ื™ืชื ื•ืช.
07:18
Will they be able to --
116
438518
1835
ื”ืื ื”ืŸ ื™ื•ื›ืœื• ืœ --
07:20
What will be their range of actions?
117
440895
2211
ืžื” ื™ื”ื™ื” ื˜ื•ื•ื— ื”ืคืขื•ืœื” ืฉืœื”ืŸ?
07:23
Will they remain limited to very specific tasks,
118
443147
3462
ื”ืื ื”ืŸ ื™ื™ืฉืืจื• ืžื•ื’ื‘ืœื•ืช ืœืžืฉื™ืžื•ืช ืกืคืฆื™ืคื™ื•ืช,
07:26
or will they be able to become a true companion across your day
119
446651
4838
ืื• ืฉื”ืŸ ื™ื•ื›ืœื• ืœื”ืคื•ืš ืœืฉื•ืชืคื•ืช ืืžืชื™ื•ืช ื‘ื™ื•ื ื™ื•ื
07:31
to which you can ask whatever you want,
120
451531
2169
ืฉืชื•ื›ืœื• ืœืฉืื•ืœ ืื•ืชืŸ ื›ืœ ืžื” ืฉืชืจืฆื•,
07:33
whatever you need?
121
453741
1210
ืœื›ืœ ืžื” ืฉืชืฆื˜ืจื›ื•?
07:35
For example, taking notes in a meeting
122
455284
2420
ืœื“ื•ื’ืžื”, ืœืจืฉื•ื ื”ืขืจื•ืช ื‘ืคื’ื™ืฉื”
07:37
or reordering milk
123
457745
2127
ืื• ืœื”ื–ืžื™ืŸ ื—ืœื‘
07:39
or even mental health coaching.
124
459872
2086
ืื• ืืคื™ืœื• ื”ื“ืจื›ื” ื‘ื‘ืจื™ืื•ืช ื”ื ืคืฉ.
07:43
Will they be able to provide you recommendations
125
463793
3045
ื”ืื ื”ืŸ ื™ื•ื›ืœื• ืœื”ืขื ื™ืง ืœืš ื”ืžืœืฆื•ืช
07:46
across product categories?
126
466838
2460
ืœื’ื‘ื™ ืžื’ื•ื•ืŸ ืงื˜ื’ื•ืจื™ื•ืช?
07:49
Today,
127
469716
1209
ื›ื™ื•ื,
07:50
companies provide us recommendations within one specific category,
128
470967
4546
ื—ื‘ืจื•ืช ืžืขื ื™ืงื•ืช ืœื ื• ื”ืžืœืฆื•ืช ื‘ืชื•ืš ืงื˜ื’ื•ืจื™ื” ืื—ืช ืกืคืฆื™ืคื™ืช,
07:55
for example, that can help us choose between two dresses,
129
475513
3795
ืœื“ื•ื’ืžื”, ืฉืชืขื–ื•ืจ ืœื ื• ืœื‘ื—ื•ืจ ื‘ื™ืŸ ืฉืชื™ ืฉืžืœื•ืช,
07:59
between two books.
130
479308
1585
ื‘ื™ืŸ ืฉื ื™ ืกืคืจื™ื.
08:01
In the future, the smart assistants will actually be able to help us choose
131
481436
4379
ื‘ืขืชื™ื“, ื”ืขื•ื–ืจื•ืช ื”ื—ื›ืžื•ืช ื™ืขื–ืจื• ืœื ื• ืœื‘ื—ื•ืจ
08:05
between buying a book or buying a dress.
132
485857
3837
ื‘ื™ืŸ ืงื ื™ื™ืช ืกืคืจ ืื• ืงื ื™ื™ืช ืฉืžืœื”.
08:10
So to be able to deliver this integrated and large range of recommendations,
133
490319
6549
ืื– ื›ื“ื™ ืœืืคืฉืจ ืœื”ืขื ื™ืง ืœื ื• ืฉื™ืœื•ื‘ ื–ื” ื‘ื™ืŸ ื˜ื•ื•ื— ืจื—ื‘ ืฉืœ ื”ืžืœืฆื•ืช,
08:16
tech teams behind smart assistants will need to design the right algorithms.
134
496868
6464
ืฆื•ื•ืชื™ื ื˜ื›ื ื™ื™ื ืžืื—ื•ืจื™ ืขื•ื–ืจื•ืช ื—ื›ืžื•ืช ื™ืฆื˜ืจื›ื• ืœืขืฆื‘ ืืช ื”ืืœื’ื•ืจื™ืชืžื™ื ื”ืžืชืื™ืžื™ื.
08:23
And these algorithms will need to be powerful enough
135
503791
3587
ื•ื”ืืœื’ื•ืจื™ืชืžื™ื ื”ืืœื” ื™ืฆื˜ืจื›ื• ืœื”ื™ื•ืช ื—ื–ืงื™ื ืžืกืคื™ืง
08:27
to process a myriad of data points.
136
507420
3754
ื›ื“ื™ ืœืขื‘ื“ ื›ืžื•ืช ืžื™ื“ืข ืขืฆื•ืžื”.
08:31
To identify patterns,
137
511966
1877
ื›ื“ื™ ืœื–ื”ื•ืช ื“ืคื•ืกื™ื,
08:33
to model courses of actions,
138
513885
2294
ืœื™ืฆื•ืจ ืžื•ื“ืœื™ื ืฉืœ ืคืขื•ืœื•ืช,
08:36
and also to learn from end users' feedback.
139
516220
2795
ื•ื’ื ืœืœืžื•ื“ ืžืžืฉื•ื‘ ืฉืœ ืžืฉืชืžืฉื™ื.
08:40
But, a world where smart assistants become unavoidable
140
520558
5839
ืื‘ืœ ืขื•ืœื ื‘ื• ืขื•ื–ืจื•ืช ื—ื›ืžื•ืช ื”ื•ืคื›ื•ืช ืœื‘ืœืชื™ ื ืžื ืขื•ืช
08:46
means new priorities for all companies,
141
526439
2711
ืžืฉืžืขื•ืชื• ืกื“ืจ ืขื“ื™ืคื•ืช ื—ื“ืฉ ืขื‘ื•ืจ ื›ืœ ื”ื—ื‘ืจื•ืช,
08:49
not only the smart assistant players.
142
529192
2794
ืœื ืจืง ื”ืฉื—ืงื ื™ื ืฉืœ ื”ืขื•ื–ืจื•ืช ื”ื—ื›ืžื•ืช.
08:52
Every business in the future will need to accelerate drastically
143
532445
4546
ื›ืœ ืขืกืง ื‘ืขืชื™ื“ ื™ืฆื˜ืจืš ืœื”ืื™ืฅ
08:56
on data and algorithm,
144
536991
2503
ืœื’ื‘ื™ ืžื™ื“ืข ื•ืืœื’ื•ืจื™ืชืžื™ื,
08:59
on voice-enabled interactions.
145
539535
3379
ืœื’ื‘ื™ ืคืขื•ืœื•ืช ืžื‘ื•ืฆืขื•ืช ืงื•ืœื™ืช.
09:02
And also, they will need to be entrusted by consumers
146
542955
3963
ื•ื’ื, ื”ื ื™ืฆื˜ืจื›ื• ืฉืœืงื•ื—ื•ืช ื™ืกืžื›ื• ืขืœื™ื”ื
09:06
to provide recommendations.
147
546959
2002
ื›ื“ื™ ืœื”ืขื ื™ืง ื”ืžืœืฆื•ืช.
09:09
This is what I like to summarize in three words,
148
549670
4004
ืืช ื–ื” ืื ื™ ืื•ื”ื‘ืช ืœืกื›ื ื‘-3 ืžื™ืœื™ื,
09:13
in the three imperatives that are data, tech and trust.
149
553674
4588
ื‘-3 ื”ืฆื™ื•ื•ื™ื™ื ืฉื”ื ืžื™ื“ืข, ื˜ื›ื ื•ืœื•ื’ื™ื™ื” ื•ืืžื•ืŸ.
09:19
So the moment the breadth challenge
150
559305
3754
ืื– ื‘ืจื’ืข ืฉืืชื’ืจ ื”ื”ื™ืงืฃ
09:23
and the voice challenge get solved
151
563059
2753
ื•ืืชื’ืจ ื”ืงื•ืœ ื™ื™ืคืชืจื•
09:25
will be a tipping point
152
565853
1460
ื–ื• ืชื”ื™ื” ื ืงื•ื“ืช ื”ืžืคื ื”
09:27
in smart assistant usage and adoption by consumers.
153
567355
4838
ื‘ืฉื™ืžื•ืฉ ื‘ืขื•ื–ืจื•ืช ื—ื›ืžื•ืช ื•ืื™ืžื•ืฆืŸ ืขืœ ื™ื“ื™ ืฆืจื›ื ื™ื.
09:33
Today, can you live without your smartphone?
154
573778
3337
ื›ื™ื•ื, ืืชื ื™ื›ื•ืœื™ื ืœื—ื™ื•ืช ืœืœื ื”ืกืžืืจื˜ืคื•ืŸ ืฉืœื›ื?
09:37
I assume not, right?
155
577490
1752
ืื ื™ ืžื ื™ื—ื” ืฉืœื, ื ื›ื•ืŸ?
09:39
In a few years from now,
156
579992
1502
ื‘ืขื•ื“ ืžืกืคืจ ืฉื ื™ื ืžื”ื™ื•ื,
09:41
your smart assistant will be a convenient,
157
581494
5213
ื”ืขื•ื–ืจืช ื”ื—ื›ืžื” ืฉืœื›ื ืชื”ื™ื” ืขื•ื–ืจืช ื ื•ื—ื”,
09:46
powerful, reliable helper
158
586749
3003
ื—ื–ืงื”, ื•ืืžื™ื ื”
09:49
essential in your day-to-day life.
159
589752
3587
ื‘ื—ื™ื™ ื”ื™ื•ื ื™ื•ื ืฉืœื›ื.
09:53
So you won't be able to live without it.
160
593339
2377
ื›ืš ืฉืœื ืชื•ื›ืœื• ืœื—ื™ื•ืช ื‘ืœืขื“ื™ื”.
09:56
And unlike your smartphones,
161
596134
2460
ื•ื‘ื ื™ื’ื•ื“ ืœืกืžืืจื˜ืคื•ื ื™ื ืฉืœื›ื,
09:58
it will be embedded in every device around you.
162
598636
3378
ื–ื” ื™ื”ื™ื” ืžื•ื‘ื ื” ื‘ื›ืœ ืžื›ืฉื™ืจ ืกื‘ื™ื‘ื›ื.
10:02
Your smartphone itself, of course,
163
602056
1919
ื‘ืกืžืืจื˜ืคื•ืŸ ืฉืœื›ื ืขืฆืžื•, ื›ืžื•ื‘ืŸ,
10:04
but also your car, the mirrors,
164
604016
3212
ืืš ื’ื ื‘ืจื›ื‘ ืฉืœื›ื, ื‘ืžืจืื•ืช,
10:07
your fridge, your glasses
165
607270
2460
ื‘ืžืงืจืจ, ื‘ืžืฉืงืคื™ื™ื
10:09
and who knows what other device in the future.
166
609730
2920
ื•ืžื™ ื™ื•ื“ืข ื‘ืื™ืœื• ืขื•ื“ ืžื›ืฉื™ืจื™ื ื‘ืขืชื™ื“.
10:14
So are you ready for a smarter life?
167
614068
3504
ืื– ื”ืื ืืชื ืžื•ื›ื ื™ื ืœื—ื™ื™ื ื—ื›ืžื™ื ื™ื•ืชืจ?
10:17
VA: Yes, Karen, I am.
168
617947
1668
ืขื•ื–ืจืช ื•ื™ืจื˜ื•ืืœื™ืช: ื›ืŸ, ืงืจืŸ.
10:19
(Laughter)
169
619907
1377
(ืฆื—ื•ืง)
10:21
Thank you.
170
621284
1167
ืชื•ื“ื” ืจื‘ื”.
10:22
(Applause)
171
622493
2586
(ืžื—ื™ืื•ืช ื›ืคื™ื™ื)
ืขืœ ืืชืจ ื–ื”

ืืชืจ ื–ื” ื™ืฆื™ื’ ื‘ืคื ื™ื›ื ืกืจื˜ื•ื ื™ YouTube ื”ืžื•ืขื™ืœื™ื ืœืœื™ืžื•ื“ ืื ื’ืœื™ืช. ืชื•ื›ืœื• ืœืจืื•ืช ืฉื™ืขื•ืจื™ ืื ื’ืœื™ืช ื”ืžื•ืขื‘ืจื™ื ืขืœ ื™ื“ื™ ืžื•ืจื™ื ืžื”ืฉื•ืจื” ื”ืจืืฉื•ื ื” ืžืจื—ื‘ื™ ื”ืขื•ืœื. ืœื—ืฅ ืคืขืžื™ื™ื ืขืœ ื”ื›ืชื•ื‘ื™ื•ืช ื‘ืื ื’ืœื™ืช ื”ืžื•ืฆื’ื•ืช ื‘ื›ืœ ื“ืฃ ื•ื™ื“ืื• ื›ื“ื™ ืœื”ืคืขื™ืœ ืืช ื”ืกืจื˜ื•ืŸ ืžืฉื. ื”ื›ืชื•ื‘ื™ื•ืช ื’ื•ืœืœื•ืช ื‘ืกื ื›ืจื•ืŸ ืขื ื”ืคืขืœืช ื”ื•ื•ื™ื“ืื•. ืื ื™ืฉ ืœืš ื”ืขืจื•ืช ืื• ื‘ืงืฉื•ืช, ืื ื ืฆื•ืจ ืื™ืชื ื• ืงืฉืจ ื‘ืืžืฆืขื•ืช ื˜ื•ืคืก ื™ืฆื™ืจืช ืงืฉืจ ื–ื”.

https://forms.gle/WvT1wiN1qDtmnspy7