1
00:00:04,160 --> 00:00:07,520
Speaker 1: Get in text with technology with tech Stuff from stuff

2
00:00:07,800 --> 00:00:14,680
Speaker 1: dot com. Hey everyone, and welcome to Tech Stuff. I

3
00:00:14,720 --> 00:00:19,480
Speaker 1: am your host, Jonathan Strickland. Today I am incredibly fortunate

4
00:00:19,560 --> 00:00:22,680
Speaker 1: I have not one, but two amazing guests to talk

5
00:00:22,760 --> 00:00:25,759
Speaker 1: about our topic of discussion today, which is really going

6
00:00:25,840 --> 00:00:30,880
Speaker 1: to be about Amazon's Alexa service and what it can do,

7
00:00:31,120 --> 00:00:34,080
Speaker 1: what what it's like to develop for it, and why

8
00:00:34,120 --> 00:00:38,880
Speaker 1: should we be excited about these voice recognition services in general. Uh.

9
00:00:39,159 --> 00:00:41,640
Speaker 1: I guarantee you by the end of this episode you'll

10
00:00:41,640 --> 00:00:44,400
Speaker 1: be as excited about them as I already am. So

11
00:00:44,479 --> 00:00:47,920
Speaker 1: let me introduce my guests. First, on the phone, I've

12
00:00:47,960 --> 00:00:51,960
Speaker 1: got Dave Zbitski from Amazon High Dave, Hey, how you

13
00:00:51,960 --> 00:00:55,880
Speaker 1: doing great? Thank you so much for joining me, really appreciated.

14
00:00:56,280 --> 00:00:59,920
Speaker 1: And I here in the studio in the flesh. John

15
00:01:00,120 --> 00:01:02,760
Speaker 1: Scheme from big nerd Ranch. Now, Josh, you've been working

16
00:01:02,840 --> 00:01:06,720
Speaker 1: on developing sort of the the the how to guide

17
00:01:06,920 --> 00:01:10,160
Speaker 1: of developing for Amazon's Alexa. Is that correct? Yeah, that's right.

18
00:01:10,160 --> 00:01:13,160
Speaker 1: I've been working with David and others in Amazon to

19
00:01:13,480 --> 00:01:17,000
Speaker 1: build some developer education tools to write apps for the

20
00:01:17,040 --> 00:01:20,080
Speaker 1: Alexa Skills Kit platform with Now, this is great. I'm

21
00:01:20,120 --> 00:01:22,679
Speaker 1: glad that I've got two experts on the subject here

22
00:01:22,760 --> 00:01:26,520
Speaker 1: because I do lots of research and I love to

23
00:01:26,600 --> 00:01:30,120
Speaker 1: chat about technology, and I'm very passionate about the subject.

24
00:01:30,600 --> 00:01:34,160
Speaker 1: But it's always a great thrill for me to have

25
00:01:34,440 --> 00:01:37,319
Speaker 1: experts on the subject matter here as well, so that

26
00:01:37,360 --> 00:01:40,160
Speaker 1: they can even fill in those gaps that are within

27
00:01:40,200 --> 00:01:44,319
Speaker 1: my understanding, because I'm coming from a consumer standpoint primarily,

28
00:01:44,360 --> 00:01:48,720
Speaker 1: I am not someone who has had a deep education

29
00:01:49,000 --> 00:01:52,000
Speaker 1: in the field of coding and developing for this sort

30
00:01:52,040 --> 00:01:57,000
Speaker 1: of stuff. I have a liberal arts degree, but I'm

31
00:01:57,120 --> 00:01:58,920
Speaker 1: so glad to have you guys here to talk about.

32
00:01:59,000 --> 00:02:01,559
Speaker 1: Let's start at the very top and work our way down.

33
00:02:02,200 --> 00:02:06,360
Speaker 1: So Alexa is uh a kind of a personal assistant

34
00:02:06,400 --> 00:02:08,720
Speaker 1: that can do lots of different stuff and depends heavily

35
00:02:08,800 --> 00:02:14,440
Speaker 1: upon voice recognition, speech recognition, natural language processing. And I

36
00:02:14,520 --> 00:02:18,320
Speaker 1: don't think a lot of people have a true understanding

37
00:02:18,400 --> 00:02:23,200
Speaker 1: or appreciation of how big a deal that is that

38
00:02:23,600 --> 00:02:26,800
Speaker 1: the way we humans communicate and the way that computers

39
00:02:26,880 --> 00:02:31,000
Speaker 1: quote unquote think is very different. So, Dave, can you

40
00:02:31,040 --> 00:02:35,079
Speaker 1: talk a little bit about the challenges of developing something

41
00:02:35,120 --> 00:02:41,799
Speaker 1: that can actually work with natural language? Certainly, I um

42
00:02:41,840 --> 00:02:43,920
Speaker 1: and and thank you for the intro. I uh, I

43
00:02:44,000 --> 00:02:46,560
Speaker 1: love talking about technology too, and I am far from

44
00:02:46,560 --> 00:02:49,440
Speaker 1: an expert. This has been a learning journey for me.

45
00:02:49,520 --> 00:02:52,960
Speaker 1: There are people who have been working on voice and

46
00:02:53,120 --> 00:02:56,800
Speaker 1: natural language and AI for thirty plus years, and I

47
00:02:56,840 --> 00:03:00,360
Speaker 1: feel like we're in a see change now with the

48
00:03:00,400 --> 00:03:06,400
Speaker 1: power of the cloud and just how affordable you know everybody, Um,

49
00:03:06,600 --> 00:03:08,360
Speaker 1: you may have a tablet or you may have a

50
00:03:08,360 --> 00:03:11,880
Speaker 1: smart home. Just that technology has become so affordable for

51
00:03:11,919 --> 00:03:15,040
Speaker 1: everybody that we can finally do something like this. And

52
00:03:15,800 --> 00:03:21,680
Speaker 1: Amazon's vision is basically the Star Trek computer. If everybody remembers,

53
00:03:22,160 --> 00:03:25,080
Speaker 1: it doesn't matter if it's you know, the Next Generation,

54
00:03:25,160 --> 00:03:29,000
Speaker 1: the original series, Deep Space nine, any or the movies, right,

55
00:03:29,320 --> 00:03:33,040
Speaker 1: there was always a voice that a human being could

56
00:03:33,120 --> 00:03:36,360
Speaker 1: just call out to in the air. I remember watching

57
00:03:36,440 --> 00:03:41,000
Speaker 1: Next Generation and uh wharf would go and play opera, right,

58
00:03:41,120 --> 00:03:44,040
Speaker 1: klingon opera or Pricard would would ask for music just

59
00:03:44,480 --> 00:03:46,520
Speaker 1: and they would walk in the room. And I was

60
00:03:46,560 --> 00:03:48,960
Speaker 1: just thinking about that the other day walking into my office,

61
00:03:49,000 --> 00:03:52,680
Speaker 1: I was like, gosh, I'm asking a computer to turn

62
00:03:52,720 --> 00:03:54,560
Speaker 1: on the lights and play a piece of music. This

63
00:03:54,720 --> 00:03:58,400
Speaker 1: is we are living in science fiction, right, And so

64
00:03:58,440 --> 00:04:02,480
Speaker 1: that's the basic idea behind mind Alexa is it is

65
00:04:02,520 --> 00:04:06,400
Speaker 1: a service from Amazon that we make available to anyone

66
00:04:06,760 --> 00:04:10,640
Speaker 1: were free, and we've put it into actual pieces of

67
00:04:10,680 --> 00:04:14,119
Speaker 1: hardware that we make and we sell to Amazon customers

68
00:04:14,160 --> 00:04:17,080
Speaker 1: as well. And some of your listeners may know that

69
00:04:17,200 --> 00:04:20,720
Speaker 1: as the Amazon Echo, the Dot, the tap, or the

70
00:04:20,760 --> 00:04:25,000
Speaker 1: fire TV. And the basic premise is that Alexa can

71
00:04:25,160 --> 00:04:29,159
Speaker 1: understand us as a human being, and then she can

72
00:04:29,160 --> 00:04:32,440
Speaker 1: talk to all the technology in our lives so that

73
00:04:32,480 --> 00:04:35,440
Speaker 1: we don't have to learn that user interface, you know,

74
00:04:35,520 --> 00:04:38,160
Speaker 1: the whole twelve o'clock blinking light. How do I go

75
00:04:38,200 --> 00:04:40,560
Speaker 1: ahead and change that. I should be able to just

76
00:04:40,760 --> 00:04:45,200
Speaker 1: ask the device itself to set the time. I should

77
00:04:45,240 --> 00:04:47,159
Speaker 1: not have to worry about that as a human being.

78
00:04:47,240 --> 00:04:51,120
Speaker 1: And that's really what Alexa is all about. She can

79
00:04:51,240 --> 00:04:55,080
Speaker 1: talk to human beings, she can understand human beings, and

80
00:04:55,400 --> 00:04:58,159
Speaker 1: then she can go ahead and tell machines what we're

81
00:04:58,200 --> 00:05:01,359
Speaker 1: actually asking for. The is really important to have that

82
00:05:01,440 --> 00:05:04,760
Speaker 1: sort of translator between us and the machine world. One

83
00:05:04,800 --> 00:05:08,680
Speaker 1: of the things that I found extremely frustrating early in

84
00:05:08,760 --> 00:05:13,719
Speaker 1: the era of home automation is that it wasn't all

85
00:05:13,839 --> 00:05:16,479
Speaker 1: or nothing kind of approach, depending upon what you wanted

86
00:05:16,480 --> 00:05:20,040
Speaker 1: to do you had to essentially go with one provider

87
00:05:20,400 --> 00:05:23,320
Speaker 1: for everything, one manufacturer for all of your stuff, because

88
00:05:23,320 --> 00:05:26,280
Speaker 1: it wouldn't talk to each other. You have different protocols,

89
00:05:26,560 --> 00:05:30,000
Speaker 1: you have different approaches to the way that they would

90
00:05:30,080 --> 00:05:32,520
Speaker 1: integrate with each other. And if you had everything from

91
00:05:32,600 --> 00:05:37,480
Speaker 1: one company, awesome, everything talks to each other, it's fantastic.

92
00:05:37,520 --> 00:05:41,040
Speaker 1: But if you're like a regular human being who can't

93
00:05:41,040 --> 00:05:44,760
Speaker 1: necessarily outfit an entire home all at once with the

94
00:05:44,880 --> 00:05:48,400
Speaker 1: same sort of technologies, you wanted to be able to

95
00:05:48,400 --> 00:05:50,640
Speaker 1: talk to each other. So I think one of the

96
00:05:50,720 --> 00:05:54,600
Speaker 1: big in my mind, one of the big bonuses of

97
00:05:54,680 --> 00:05:58,200
Speaker 1: an approach like Alexa is the idea that you have

98
00:05:58,400 --> 00:06:01,880
Speaker 1: this go between that can do that work for you

99
00:06:02,120 --> 00:06:05,800
Speaker 1: where it can start to compensate for the fact that

100
00:06:05,839 --> 00:06:10,599
Speaker 1: these technologies don't natively talk to each other necessarily. Um. Right.

101
00:06:10,680 --> 00:06:14,520
Speaker 1: A great thing about standards is there's so many of them, right, yeah, yeah.

102
00:06:14,520 --> 00:06:17,080
Speaker 1: I love the fact that you know, the term standard

103
00:06:17,320 --> 00:06:19,520
Speaker 1: means the opposite of what you could expect it to

104
00:06:19,600 --> 00:06:23,440
Speaker 1: be and um and and that is one of the

105
00:06:23,440 --> 00:06:27,200
Speaker 1: most popular uses we have seen for the the echo device.

106
00:06:27,279 --> 00:06:29,640
Speaker 1: I know that was my journey too, is I started

107
00:06:29,680 --> 00:06:32,640
Speaker 1: out you know. I used it for music and general queries,

108
00:06:32,640 --> 00:06:35,600
Speaker 1: and I said, you know, I heard this thing can

109
00:06:35,640 --> 00:06:37,880
Speaker 1: do stuff with smart home. I have no idea what

110
00:06:37,960 --> 00:06:39,960
Speaker 1: smart home is. I have no idea what any of

111
00:06:39,960 --> 00:06:43,520
Speaker 1: these terms like IoT right, internet things. It doesn't make

112
00:06:43,560 --> 00:06:45,680
Speaker 1: sense to me. I want to light I can turn on.

113
00:06:46,080 --> 00:06:49,000
Speaker 1: So I went to Amazon and I searched for smart

114
00:06:49,080 --> 00:06:51,120
Speaker 1: light bulb I think is what I found, and it

115
00:06:51,200 --> 00:06:54,320
Speaker 1: was I think it was fifteen bucks. I figured i'd

116
00:06:54,400 --> 00:06:56,200
Speaker 1: order it and see if it would work. And there

117
00:06:56,279 --> 00:06:59,640
Speaker 1: was you know, kind of a general instruction on Amazon site.

118
00:07:00,080 --> 00:07:02,960
Speaker 1: But what you can actually do with Alexa as you

119
00:07:02,960 --> 00:07:06,400
Speaker 1: get the device and you say, Alexa discover devices and

120
00:07:06,400 --> 00:07:08,640
Speaker 1: then she figures out what you put into your home.

121
00:07:08,720 --> 00:07:11,080
Speaker 1: You don't have to go and figure that out. And

122
00:07:11,120 --> 00:07:15,040
Speaker 1: we've found enough custard demand that we've actually created. If

123
00:07:15,040 --> 00:07:19,200
Speaker 1: you go to Amazon dot com slash smart Home, people

124
00:07:19,240 --> 00:07:21,559
Speaker 1: can go there and they'll see all of the smart

125
00:07:21,600 --> 00:07:24,760
Speaker 1: home devices that Alexa can just talk to and will

126
00:07:24,800 --> 00:07:27,880
Speaker 1: automatically work. So it makes it a much easier process

127
00:07:28,320 --> 00:07:31,520
Speaker 1: versus trying to figure out all the individual pieces to buy,

128
00:07:31,600 --> 00:07:33,760
Speaker 1: who talks to who and everything that you mentioned. You

129
00:07:33,760 --> 00:07:36,560
Speaker 1: can just go ahead and put a light screw in

130
00:07:36,600 --> 00:07:38,280
Speaker 1: a light bulb and be able to talk to it. No,

131
00:07:39,360 --> 00:07:42,240
Speaker 1: that's incredible from a consumer standpoint, right, the idea that

132
00:07:42,280 --> 00:07:46,400
Speaker 1: you've made this sort of a seamless experience so that

133
00:07:46,800 --> 00:07:50,600
Speaker 1: you don't have that frustrating moment where even with something

134
00:07:50,640 --> 00:07:53,320
Speaker 1: as simple as bluetooth pairing, for some people that that

135
00:07:53,520 --> 00:07:57,080
Speaker 1: is that is a barrier, right that they have to wait,

136
00:07:57,120 --> 00:07:59,600
Speaker 1: do I have to on each device when the light

137
00:07:59,680 --> 00:08:02,440
Speaker 1: is bling? Game? It means that what's happening, you know,

138
00:08:02,600 --> 00:08:04,880
Speaker 1: just something as simple as that can be really frustrating

139
00:08:04,920 --> 00:08:08,080
Speaker 1: for some people. So to take that step away, uh,

140
00:08:08,240 --> 00:08:12,560
Speaker 1: is really really an ingenious and helpful thing to do. Now, Now, Josh,

141
00:08:13,040 --> 00:08:16,440
Speaker 1: you've worked very hard to help with the back end

142
00:08:16,480 --> 00:08:19,360
Speaker 1: of this so that people who are developing for Alexa

143
00:08:19,760 --> 00:08:23,920
Speaker 1: can take advantage of this and and give people chance

144
00:08:24,000 --> 00:08:28,239
Speaker 1: to make Alexa do some pretty incredible things. So, first

145
00:08:28,280 --> 00:08:32,640
Speaker 1: of all, I gotta I gotta lay down some vocabulary.

146
00:08:32,800 --> 00:08:37,040
Speaker 1: Right before we started recording, you talked about how a

147
00:08:37,120 --> 00:08:39,079
Speaker 1: thing on your phone you refer to it as a skill,

148
00:08:39,080 --> 00:08:41,439
Speaker 1: and you think all of us, because I'm so deep

149
00:08:41,480 --> 00:08:44,760
Speaker 1: in this Amazon world, skills and acts. So so tell

150
00:08:44,800 --> 00:08:49,000
Speaker 1: people what what exactly is an Alexa skill? Well, it's

151
00:08:49,000 --> 00:08:53,160
Speaker 1: interesting skill the term you think initially a skill is

152
00:08:53,200 --> 00:08:56,160
Speaker 1: something that you acquire or learn over time. And I

153
00:08:56,200 --> 00:08:59,320
Speaker 1: believe that Amazon used that terminology as a nod towards

154
00:09:00,000 --> 00:09:02,880
Speaker 1: really the machine learning aspect of what they offer in

155
00:09:02,920 --> 00:09:06,120
Speaker 1: the cloud, right, um, which is and David was speaking

156
00:09:06,160 --> 00:09:08,959
Speaker 1: to this a bit ago. Um, you know, the machine

157
00:09:09,040 --> 00:09:13,640
Speaker 1: learning component of what the Amazon service offers, and that

158
00:09:13,840 --> 00:09:19,280
Speaker 1: is the incredibly complex problem, right of taking spoken words

159
00:09:19,440 --> 00:09:23,640
Speaker 1: and resolving them to a format that uh software can

160
00:09:23,679 --> 00:09:27,880
Speaker 1: actually work with and treat in a predictable way. Um.

161
00:09:27,920 --> 00:09:32,199
Speaker 1: You have to imagine all the timing, the inflection, the variation,

162
00:09:32,320 --> 00:09:37,320
Speaker 1: just regional differences, uh, that someone's going to ask for something.

163
00:09:37,360 --> 00:09:39,400
Speaker 1: And then also you have to think about all the

164
00:09:39,480 --> 00:09:42,280
Speaker 1: various ways that someone could ask for the exact same things.

165
00:09:42,440 --> 00:09:45,840
Speaker 1: It's an astronomical number of ways that that can happen.

166
00:09:47,000 --> 00:09:50,640
Speaker 1: And it's the Alexa Skills Kit platform that actually resolves

167
00:09:50,679 --> 00:09:54,040
Speaker 1: that problem, um in a lot of ways, UM for you.

168
00:09:54,240 --> 00:09:57,000
Speaker 1: So I believe that's why they chose that terminology. It

169
00:09:57,040 --> 00:10:01,280
Speaker 1: makes sense, right, And actually Amazon's corporate a corpus by

170
00:10:01,320 --> 00:10:04,600
Speaker 1: the way, to throw another technology steward into the mix.

171
00:10:04,800 --> 00:10:08,760
Speaker 1: UM is really a collection of data, and Amazon their

172
00:10:08,800 --> 00:10:12,960
Speaker 1: service offers a large collection of that data at which

173
00:10:13,000 --> 00:10:16,200
Speaker 1: is ever increasing by the way. UM. That simplifies the

174
00:10:16,240 --> 00:10:20,319
Speaker 1: problem of resolving that speech and converting it into UM.

175
00:10:20,400 --> 00:10:23,680
Speaker 1: What what the platform calls an intent, which is really

176
00:10:23,679 --> 00:10:25,880
Speaker 1: an indication of something that someone would like to do

177
00:10:25,960 --> 00:10:29,240
Speaker 1: at a given time. Yeah, it's interesting because if you

178
00:10:29,280 --> 00:10:32,440
Speaker 1: think about it from a classical computing standpoint, if you

179
00:10:32,480 --> 00:10:36,040
Speaker 1: wanted your computer to do something specific, uh, let's go

180
00:10:36,080 --> 00:10:37,840
Speaker 1: back to the DOS days. We're not gonna go all

181
00:10:37,880 --> 00:10:39,440
Speaker 1: the way back. We'll go back to DOST days because

182
00:10:39,440 --> 00:10:42,240
Speaker 1: that's that's my childhood. Then you would type in a

183
00:10:42,280 --> 00:10:45,600
Speaker 1: command and and the computer knew exactly what you wanted

184
00:10:45,640 --> 00:10:48,520
Speaker 1: to do as long as the actual program is installed

185
00:10:48,520 --> 00:10:52,520
Speaker 1: on your computer. Because you're following a very specific protocol

186
00:10:52,559 --> 00:10:55,160
Speaker 1: that does not vary. That's right, always going to be

187
00:10:55,200 --> 00:10:56,760
Speaker 1: the same. It's a one to one thing, and it's

188
00:10:56,800 --> 00:11:00,240
Speaker 1: a textual interface and very consistent. Yes, but when you

189
00:11:00,240 --> 00:11:03,120
Speaker 1: get to two different people, just just two people, and

190
00:11:03,160 --> 00:11:05,719
Speaker 1: you just want them to ask for the same thing,

191
00:11:05,800 --> 00:11:09,320
Speaker 1: but you're not guiding them in how to ask, that's

192
00:11:09,320 --> 00:11:12,040
Speaker 1: same for that same thing. That's where you start getting

193
00:11:12,080 --> 00:11:16,920
Speaker 1: into this, uh, this this variability and even if they're

194
00:11:16,920 --> 00:11:19,720
Speaker 1: both saying the exact same phrase, if they're from my

195
00:11:19,840 --> 00:11:21,599
Speaker 1: neck of the woods, there might be a bit of

196
00:11:21,640 --> 00:11:24,600
Speaker 1: a droll that's right there. If they're up over in Maine,

197
00:11:24,679 --> 00:11:26,559
Speaker 1: it's going to be a different sound if they're if

198
00:11:26,559 --> 00:11:28,800
Speaker 1: they're no non native English speaker, they're gonna be in

199
00:11:29,000 --> 00:11:32,480
Speaker 1: an inflection in their voice from whatever language was their

200
00:11:32,480 --> 00:11:36,400
Speaker 1: primary language. So these are all non trivial problems actually

201
00:11:36,920 --> 00:11:39,679
Speaker 1: in the in the programming world. Yes, so there's that problem,

202
00:11:39,800 --> 00:11:42,360
Speaker 1: and then there's also an Amazon refers to this as

203
00:11:42,440 --> 00:11:46,400
Speaker 1: the interaction model, which is the number of variances in

204
00:11:46,480 --> 00:11:50,200
Speaker 1: how someone could ask for something, and the platform takes

205
00:11:50,280 --> 00:11:53,880
Speaker 1: a fuzzy matching approach to solving that problem. Right, So

206
00:11:54,400 --> 00:11:57,440
Speaker 1: instead of providing an exhaustive list of all of the

207
00:11:57,520 --> 00:12:01,320
Speaker 1: various ways that someone could ask for information for an airport,

208
00:12:01,440 --> 00:12:06,880
Speaker 1: for example, um, instead they use a artificial intelligence approach

209
00:12:06,880 --> 00:12:10,840
Speaker 1: to that problem and generalize a set of training data

210
00:12:10,880 --> 00:12:14,640
Speaker 1: that you actually provide as a developer, uh to to

211
00:12:14,679 --> 00:12:18,120
Speaker 1: simplify that problem down. Okay, so yeah, because the first

212
00:12:18,160 --> 00:12:19,640
Speaker 1: thing I was guessing was like I wonder if it's

213
00:12:19,679 --> 00:12:21,480
Speaker 1: going to be probabilistic. It is one of those things

214
00:12:21,480 --> 00:12:24,560
Speaker 1: where it assigns a probability that I'm pretty sure this

215
00:12:24,600 --> 00:12:26,760
Speaker 1: is what they're asking for, so let's go for that.

216
00:12:26,920 --> 00:12:29,720
Speaker 1: Uh exactly. Yeah. We've talked about that with some of

217
00:12:29,760 --> 00:12:33,440
Speaker 1: the other artificial intelligence platforms out there, things like IBM

218
00:12:33,480 --> 00:12:36,720
Speaker 1: S Watson being a very simple example, right, simple in

219
00:12:36,720 --> 00:12:39,560
Speaker 1: the sense that it's easy to understand. It's actually a

220
00:12:39,600 --> 00:12:42,760
Speaker 1: pretty complicated machine as it turns out. But the fact

221
00:12:42,840 --> 00:12:44,559
Speaker 1: that you would say, all right, well, when it was

222
00:12:44,559 --> 00:12:47,679
Speaker 1: playing jeopardy, it would never buzz in unless that certainty

223
00:12:47,760 --> 00:12:50,920
Speaker 1: was greater than like an eight threshold. And once you

224
00:12:51,000 --> 00:12:53,959
Speaker 1: explain to people that's what we mean by probabilistic, where

225
00:12:54,040 --> 00:12:58,679
Speaker 1: a computer is determining, well, how how quote unquote sure

226
00:12:58,800 --> 00:13:01,880
Speaker 1: am I that this is intent of the person giving

227
00:13:01,920 --> 00:13:05,080
Speaker 1: the command then act upon it. One of the things

228
00:13:05,080 --> 00:13:08,160
Speaker 1: I wanted to also mention about alexas skill, so we

229
00:13:08,320 --> 00:13:11,760
Speaker 1: talked about how it's it's voice commands, right that you're

230
00:13:11,760 --> 00:13:16,360
Speaker 1: giving it. These have their own kind of anatomy, Right,

231
00:13:16,440 --> 00:13:20,320
Speaker 1: You've got the skill, where how how you activate the

232
00:13:20,320 --> 00:13:23,240
Speaker 1: skill itself? What you call upon in order to make

233
00:13:23,240 --> 00:13:26,280
Speaker 1: the skill happen, and UM, I wonder if you can

234
00:13:26,320 --> 00:13:28,560
Speaker 1: maybe go into just a little bit like it might

235
00:13:28,559 --> 00:13:30,959
Speaker 1: not be something that the end user would necessarily think

236
00:13:30,960 --> 00:13:33,839
Speaker 1: about the developers would think about. Yeah, no exactly that.

237
00:13:34,000 --> 00:13:36,160
Speaker 1: You know, it's got its own grab bag of terminology,

238
00:13:36,240 --> 00:13:39,040
Speaker 1: but it's really not that complicated once you get past

239
00:13:39,040 --> 00:13:42,240
Speaker 1: those initial things. Um. You know. So as a developer

240
00:13:42,320 --> 00:13:45,079
Speaker 1: coming to the platform, the first thing that I asked was, well,

241
00:13:45,120 --> 00:13:47,560
Speaker 1: what heck is the name of my app or skill?

242
00:13:47,679 --> 00:13:50,800
Speaker 1: I mean, um and updating that in my mind? UM,

243
00:13:50,840 --> 00:13:53,600
Speaker 1: and that is called the invocation name on the platform.

244
00:13:53,720 --> 00:13:57,240
Speaker 1: So what that invocation does is it invocation name does?

245
00:13:57,320 --> 00:13:59,920
Speaker 1: Is it maps a user's word? Basically? I think it

246
00:14:00,120 --> 00:14:02,480
Speaker 1: as a name space like I came to you know,

247
00:14:02,960 --> 00:14:06,600
Speaker 1: Alexa skills kit development as a Java developer and being

248
00:14:06,600 --> 00:14:10,520
Speaker 1: able to give package names or name spaces two classes

249
00:14:10,840 --> 00:14:14,319
Speaker 1: as a Java developer is very helpful. Um. And really

250
00:14:14,360 --> 00:14:18,439
Speaker 1: that's what it does, uh, an invocation named. So for example, UM,

251
00:14:18,480 --> 00:14:21,560
Speaker 1: in our class at Bigner Ranch that we've built for

252
00:14:21,560 --> 00:14:25,120
Speaker 1: for Amazon, we we've given We've built a skill to

253
00:14:25,280 --> 00:14:29,280
Speaker 1: give you information about airports and UM, you you say, Alexa,

254
00:14:29,400 --> 00:14:32,200
Speaker 1: ask airport info for flight delays at a t L.

255
00:14:32,280 --> 00:14:36,440
Speaker 1: For example, that airport info word or words is an

256
00:14:36,440 --> 00:14:39,600
Speaker 1: invocation name, so that initially brings up the skill. It

257
00:14:39,680 --> 00:14:42,440
Speaker 1: launches it so to speak UM. And then you've got

258
00:14:43,040 --> 00:14:46,840
Speaker 1: sample utterances UM. I mentioned that you train UM. I

259
00:14:47,240 --> 00:14:49,120
Speaker 1: kind of think of it as a brain in the cloud,

260
00:14:49,360 --> 00:14:51,360
Speaker 1: UM that lives in the cloud to do our bidding

261
00:14:51,400 --> 00:14:55,240
Speaker 1: for us for resolving the spoken words. UH. These sample

262
00:14:55,320 --> 00:14:59,080
Speaker 1: utterances are what effectively resolve what someone's asking for to

263
00:14:59,720 --> 00:15:02,840
Speaker 1: an indication or an intent that we'd like to ask

264
00:15:02,880 --> 00:15:07,080
Speaker 1: for airport information. And Amazon has made some really smart

265
00:15:07,160 --> 00:15:12,000
Speaker 1: decisions about making that a black box. Effectively, as a developer,

266
00:15:12,440 --> 00:15:15,000
Speaker 1: all we do is provide that training data and on

267
00:15:15,040 --> 00:15:17,600
Speaker 1: the other side we get an indication of what came out.

268
00:15:18,040 --> 00:15:21,360
Speaker 1: We're not required to you know, set up a machine

269
00:15:21,480 --> 00:15:24,800
Speaker 1: learning server or you know, artificial intelligence or deal with

270
00:15:24,840 --> 00:15:28,440
Speaker 1: any of those algorithms that who knows how many countless

271
00:15:28,840 --> 00:15:33,600
Speaker 1: engineers and hours Amazon is invested into building that infrastructure UM.

272
00:15:33,640 --> 00:15:36,360
Speaker 1: But we've got that as a tool to resolve the

273
00:15:36,360 --> 00:15:40,520
Speaker 1: information down to something that our skill service can work with.

274
00:15:40,720 --> 00:15:43,680
Speaker 1: So with an Alexis skill, you've got a skill interface

275
00:15:43,720 --> 00:15:45,960
Speaker 1: and a skill service the Skill interface is where this

276
00:15:46,040 --> 00:15:49,000
Speaker 1: brain that I keep mentioning lives, and the skill service

277
00:15:49,200 --> 00:15:53,560
Speaker 1: is it's really anything that can speak HTTPS. UM. Now

278
00:15:54,000 --> 00:15:57,240
Speaker 1: with our class, we're using no JS, which is, you know,

279
00:15:57,280 --> 00:16:00,240
Speaker 1: it's JavaScript on the server side. Everybody's it in a

280
00:16:00,360 --> 00:16:03,520
Speaker 1: dab of JavaScript here there anyway, So it's an easy

281
00:16:03,560 --> 00:16:05,880
Speaker 1: language for people to pick up and get into. If

282
00:16:05,880 --> 00:16:09,360
Speaker 1: you've done you know, um uh, any web development, you've

283
00:16:09,400 --> 00:16:13,480
Speaker 1: likely got some JavaScript exposure. UM. So it's really an

284
00:16:13,520 --> 00:16:16,520
Speaker 1: extension upon that using some you know, you've got some

285
00:16:16,560 --> 00:16:19,600
Speaker 1: additional things like file io and being able to write

286
00:16:19,640 --> 00:16:23,480
Speaker 1: to a database using node. Uh. But with that JavaScript layer,

287
00:16:23,760 --> 00:16:26,160
Speaker 1: we get events from the Skill interface and we're able

288
00:16:26,200 --> 00:16:29,560
Speaker 1: to process those events and then send a message back

289
00:16:29,680 --> 00:16:34,120
Speaker 1: using Jason right Java JavaScript object notation uh to the

290
00:16:34,280 --> 00:16:37,480
Speaker 1: to the device, and she's able to speak speak our response.

291
00:16:37,640 --> 00:16:40,400
Speaker 1: It's pretty cool. This is really well, I'm sorry, Dave,

292
00:16:40,440 --> 00:16:43,880
Speaker 1: please go ahead. I was just gonna say to um

293
00:16:43,920 --> 00:16:46,160
Speaker 1: to to add on to that. You can think of

294
00:16:46,200 --> 00:16:49,440
Speaker 1: it in terms of you know, we've we talked before

295
00:16:49,520 --> 00:16:54,360
Speaker 1: about Alexa translating human language into a way that she

296
00:16:54,440 --> 00:16:58,800
Speaker 1: can talk to technology. It is the same when third

297
00:16:58,960 --> 00:17:02,280
Speaker 1: party developers go ahead and enhance what she can understand

298
00:17:02,760 --> 00:17:07,040
Speaker 1: through skills. So, if you are an Amazon customer and

299
00:17:07,160 --> 00:17:10,440
Speaker 1: you have an Echo and you talk to your echo,

300
00:17:11,000 --> 00:17:14,040
Speaker 1: and perhaps you're using what Josh describeder, you're using a

301
00:17:14,119 --> 00:17:18,680
Speaker 1: starting phrase, uh and a invocation name to launch something

302
00:17:18,760 --> 00:17:22,480
Speaker 1: like let's say Fitbit or uber or your your Domino's

303
00:17:22,520 --> 00:17:26,359
Speaker 1: easy order. Your voice goes to Amazon. Your voice is

304
00:17:26,440 --> 00:17:29,359
Speaker 1: never shared with a third party developer. You should think

305
00:17:29,440 --> 00:17:32,800
Speaker 1: of Alexa almost as that friend that can can not

306
00:17:32,880 --> 00:17:35,840
Speaker 1: only translate to other pieces of technology, but she can

307
00:17:35,880 --> 00:17:38,960
Speaker 1: talk to all of these third party developers who have

308
00:17:39,080 --> 00:17:44,280
Speaker 1: their own technology. Uh. And so in essence, for for Fitbit,

309
00:17:45,000 --> 00:17:48,080
Speaker 1: when you talk to her, she understands that you're asking

310
00:17:48,320 --> 00:17:50,320
Speaker 1: how you know, how did you do today? And then

311
00:17:50,359 --> 00:17:53,600
Speaker 1: she goes and she talks to Fitbit, and Fitbit says, oh,

312
00:17:53,680 --> 00:17:55,119
Speaker 1: you know, we have these servers and we have all

313
00:17:55,160 --> 00:17:57,960
Speaker 1: these datas and and we know we uh, you know,

314
00:17:58,040 --> 00:17:59,920
Speaker 1: we we know you're a customer. We're gonna return that.

315
00:18:00,400 --> 00:18:03,600
Speaker 1: So go tell that customer this. And then Alexa goes

316
00:18:04,000 --> 00:18:08,000
Speaker 1: and she tells you the information that Fitbit had for you. So,

317
00:18:08,040 --> 00:18:13,040
Speaker 1: in in essence, it is translating between those things and uh,

318
00:18:13,160 --> 00:18:17,480
Speaker 1: you mentioned before the probability. Uh, it's very interesting. As

319
00:18:17,560 --> 00:18:22,040
Speaker 1: human beings when we have a conversation, we were constantly

320
00:18:22,080 --> 00:18:25,399
Speaker 1: making choices like that too. I bring all of my

321
00:18:25,480 --> 00:18:28,120
Speaker 1: experience to this podcast today, right, So when we're having

322
00:18:28,160 --> 00:18:31,280
Speaker 1: a conversation, it's based on all the years that I've

323
00:18:31,320 --> 00:18:34,280
Speaker 1: had a conversation and the understanding of what those words

324
00:18:34,800 --> 00:18:37,479
Speaker 1: actually mean. And there's a lot of things that go

325
00:18:37,600 --> 00:18:40,000
Speaker 1: on set. So for example, we all know what time

326
00:18:40,000 --> 00:18:42,320
Speaker 1: of the day it is where we are, we know

327
00:18:42,400 --> 00:18:46,000
Speaker 1: where we live, we know what country we're in. I

328
00:18:46,119 --> 00:18:48,280
Speaker 1: know you know, I know how to stand up, I

329
00:18:48,400 --> 00:18:50,600
Speaker 1: know what eyes and nose look like. I know what

330
00:18:50,640 --> 00:18:53,200
Speaker 1: a computer ist I knew what when you reference Doss,

331
00:18:54,320 --> 00:18:57,520
Speaker 1: A lot of machine learning is that a computer doesn't

332
00:18:57,560 --> 00:19:01,560
Speaker 1: necessarily have that context, right, And so when you're a

333
00:19:01,600 --> 00:19:06,119
Speaker 1: third party developer, you know the the interaction model that

334
00:19:06,240 --> 00:19:09,119
Speaker 1: jos described basically, think of it like, Uh, you and

335
00:19:09,160 --> 00:19:10,679
Speaker 1: I are going to have a discussion around a new

336
00:19:10,680 --> 00:19:13,359
Speaker 1: topic today. Maybe it's something to do with fitness, and

337
00:19:13,440 --> 00:19:15,000
Speaker 1: so we all, you know, we all go look at

338
00:19:15,000 --> 00:19:17,080
Speaker 1: a wiki or maybe we look at Reddit, and we

339
00:19:17,160 --> 00:19:19,160
Speaker 1: define all the terms and then you and I can

340
00:19:19,359 --> 00:19:22,359
Speaker 1: can have a conversation and actually understand all of that.

341
00:19:22,440 --> 00:19:25,600
Speaker 1: And so that's what you're basically setting up when you

342
00:19:25,640 --> 00:19:28,400
Speaker 1: want to add an additional skill to Alexa is Hey,

343
00:19:28,440 --> 00:19:31,600
Speaker 1: here's all the terms, and when people ask for it

344
00:19:31,680 --> 00:19:34,800
Speaker 1: like this in language, here's what I want you to

345
00:19:34,840 --> 00:19:38,720
Speaker 1: tell me that they asked. And Dave you touched upon

346
00:19:38,840 --> 00:19:41,159
Speaker 1: something early on in that too that I think is

347
00:19:41,560 --> 00:19:44,879
Speaker 1: interesting and important to point out the idea that uh

348
00:19:44,880 --> 00:19:49,440
Speaker 1: with with you communicating to Alexa, your information is going

349
00:19:49,480 --> 00:19:52,880
Speaker 1: to Amazon. It's not being propagated across the internet willy nilly,

350
00:19:53,040 --> 00:19:57,200
Speaker 1: being spread everywhere. That also ends up being an important

351
00:19:57,240 --> 00:19:59,679
Speaker 1: part of some of the restrictions around the types of

352
00:19:59,680 --> 00:20:03,359
Speaker 1: skill ales that Amazon will accept for Alexa, one being

353
00:20:03,680 --> 00:20:07,240
Speaker 1: that they there aren't going to be uh child once

354
00:20:07,400 --> 00:20:11,879
Speaker 1: once for children specifically, because there is a very real

355
00:20:11,960 --> 00:20:15,879
Speaker 1: concern about privacy and information, particularly for children who may

356
00:20:15,880 --> 00:20:20,520
Speaker 1: not fully understand or recognize the importance of that. Um.

357
00:20:20,560 --> 00:20:23,959
Speaker 1: So are there other restrictions on the types of skills

358
00:20:24,000 --> 00:20:27,880
Speaker 1: Amazon is going to except I would imagine that anything

359
00:20:27,880 --> 00:20:33,760
Speaker 1: that was against the law right out the door not happening. Yeah,

360
00:20:33,800 --> 00:20:37,879
Speaker 1: you know, um in the scenario that you described, because

361
00:20:37,880 --> 00:20:41,480
Speaker 1: Alexa today has no way to distinguish between different voices.

362
00:20:41,520 --> 00:20:44,280
Speaker 1: So if we had an echo in the room and

363
00:20:44,359 --> 00:20:47,080
Speaker 1: I was talking and you were talking, that is both

364
00:20:47,240 --> 00:20:50,679
Speaker 1: a human being, it's it's language. She cannot differentiate. So

365
00:20:50,720 --> 00:20:53,679
Speaker 1: she cannot differentiate whether it is a child asking for

366
00:20:53,760 --> 00:20:56,879
Speaker 1: something or an adult. And because of that, uh, and

367
00:20:56,960 --> 00:20:59,600
Speaker 1: you know there are COPA regulations and and everything else

368
00:20:59,640 --> 00:21:02,320
Speaker 1: when it is to mobile and web existing today, we

369
00:21:02,320 --> 00:21:04,480
Speaker 1: want to make sure that we're honoring that and that

370
00:21:04,560 --> 00:21:11,880
Speaker 1: we are protecting folks. Um your um uh second question

371
00:21:12,080 --> 00:21:15,920
Speaker 1: around um, if there's anything else that we're not necessarily allowing,

372
00:21:15,960 --> 00:21:19,600
Speaker 1: I think, um, there are. The guidelines are very similar.

373
00:21:19,680 --> 00:21:22,439
Speaker 1: So we do have an Android app store today, an

374
00:21:22,480 --> 00:21:25,359
Speaker 1: Amazon App Store which has very similar policy. So it

375
00:21:25,400 --> 00:21:28,520
Speaker 1: is something that Amazon has been in that that space

376
00:21:28,680 --> 00:21:32,119
Speaker 1: for years and we've kind of learned as we've gone through.

377
00:21:32,560 --> 00:21:35,680
Speaker 1: We just launched an update to the Alexis Skills section

378
00:21:36,640 --> 00:21:39,440
Speaker 1: where we have categories, and those categories are very similar

379
00:21:39,440 --> 00:21:42,160
Speaker 1: to what people would expect in a mobile app store

380
00:21:42,280 --> 00:21:45,800
Speaker 1: for you know, games, technology, things like that. So uh,

381
00:21:45,880 --> 00:21:48,080
Speaker 1: you know it's very it's got to be PG thirteen

382
00:21:48,160 --> 00:21:51,960
Speaker 1: and G you know, no our rated content anything like that.

383
00:21:52,640 --> 00:21:57,439
Speaker 1: No UM personally identifiable information. You would not. You know,

384
00:21:57,480 --> 00:22:00,400
Speaker 1: we're very careful about if you keep asking people all

385
00:22:00,440 --> 00:22:02,520
Speaker 1: about them, why are you asking about them? And you

386
00:22:02,560 --> 00:22:06,040
Speaker 1: better have a privacy policy for that that somebody has

387
00:22:06,080 --> 00:22:09,439
Speaker 1: allowed you to ask for that information. We do not

388
00:22:09,600 --> 00:22:14,520
Speaker 1: share any information about you. We don't share device information UM,

389
00:22:14,720 --> 00:22:18,280
Speaker 1: although you know developers of course keep asking for those

390
00:22:18,320 --> 00:22:19,960
Speaker 1: things you know they want to know is this coming

391
00:22:19,960 --> 00:22:23,360
Speaker 1: from an echo or mulbile app? Because Alexa will run

392
00:22:23,480 --> 00:22:26,600
Speaker 1: in anything. So there are UH apps for the iPhone,

393
00:22:26,640 --> 00:22:28,879
Speaker 1: apps for Android where you can talk to Alexa and

394
00:22:28,920 --> 00:22:31,399
Speaker 1: get access to all your skills. You can get access

395
00:22:31,400 --> 00:22:34,200
Speaker 1: to all of your smart home functionality as well, using

396
00:22:34,200 --> 00:22:39,480
Speaker 1: the same familiar Alexa experience. See and I am glad

397
00:22:39,520 --> 00:22:41,600
Speaker 1: you were able to address that. I think that that

398
00:22:41,760 --> 00:22:44,600
Speaker 1: actually is a very important thing for any platform to

399
00:22:44,680 --> 00:22:48,520
Speaker 1: do well. You do. Actually it seems counterintuitive, but you

400
00:22:48,600 --> 00:22:51,680
Speaker 1: do need to have those guidelines and restrictions there. And

401
00:22:52,119 --> 00:22:56,520
Speaker 1: if you ever want to see what can happen if

402
00:22:56,560 --> 00:22:59,560
Speaker 1: you don't put them there, you can look at some

403
00:23:00,000 --> 00:23:05,040
Speaker 1: any dramatic lessons we've learned through things like Microsoft Ta,

404
00:23:05,640 --> 00:23:10,240
Speaker 1: which was certainly not intended to turn into a big problem.

405
00:23:10,359 --> 00:23:13,840
Speaker 1: It was that the intention, the intent was how long

406
00:23:13,880 --> 00:23:16,480
Speaker 1: did that last? For like a day and hours? It

407
00:23:16,560 --> 00:23:18,920
Speaker 1: was twenty four hours and then they pulled the plug. Yeah,

408
00:23:18,920 --> 00:23:20,879
Speaker 1: I did a full episode of Microsoft Taste. I'm not

409
00:23:20,920 --> 00:23:23,040
Speaker 1: going to go back and do that, obviously, I'm not

410
00:23:23,080 --> 00:23:25,359
Speaker 1: going to ask either of you to comment on that

411
00:23:25,400 --> 00:23:29,240
Speaker 1: anymore than but just to just to say, like, if

412
00:23:29,280 --> 00:23:32,639
Speaker 1: you have a system and you don't have those restrictions

413
00:23:32,640 --> 00:23:37,120
Speaker 1: in and we are all in many ways like children,

414
00:23:37,240 --> 00:23:40,960
Speaker 1: and sometimes as children, you want to test boundaries, and

415
00:23:41,080 --> 00:23:45,600
Speaker 1: if you find there are no boundaries, problems happen. So

416
00:23:46,240 --> 00:23:49,520
Speaker 1: I'm in favor of boundaries personally. That's actually one thing,

417
00:23:49,840 --> 00:23:52,560
Speaker 1: uh that Amazon has done a really good job of

418
00:23:52,800 --> 00:23:58,200
Speaker 1: is curating the experience as well. You know, they literally, um,

419
00:23:58,240 --> 00:24:02,120
Speaker 1: as you submit your skill, audit that skill pretty thoroughly

420
00:24:02,119 --> 00:24:04,560
Speaker 1: and pretty rigorously to make sure that it conforms to

421
00:24:05,440 --> 00:24:09,720
Speaker 1: good Voice User Experience guidelines. Well, and when you're talking

422
00:24:09,720 --> 00:24:12,600
Speaker 1: about a device that that you know people think of

423
00:24:12,720 --> 00:24:16,640
Speaker 1: as listening to you, obviously you have a great responsibility

424
00:24:17,280 --> 00:24:20,439
Speaker 1: in order to provide an experience that isn't going to

425
00:24:20,600 --> 00:24:23,919
Speaker 1: be negative in any way. Uh, knowing that you know

426
00:24:23,960 --> 00:24:26,040
Speaker 1: you have to take a lot of time and effort

427
00:24:26,080 --> 00:24:29,560
Speaker 1: to make certain that you avoid any problems that could

428
00:24:29,600 --> 00:24:31,840
Speaker 1: come later down the line. I mean, that's got to

429
00:24:31,880 --> 00:24:36,119
Speaker 1: be pretty pretty top concern from Amazon. Yeah, that was,

430
00:24:37,080 --> 00:24:40,320
Speaker 1: you know, very very important for us at Amazon when

431
00:24:40,320 --> 00:24:43,680
Speaker 1: we created the device. And so if you're not using

432
00:24:43,680 --> 00:24:46,479
Speaker 1: an Echo, you're using another device, you know, maybe it's

433
00:24:46,480 --> 00:24:48,919
Speaker 1: a mobile app, or you know, maybe it's in the

434
00:24:49,000 --> 00:24:52,040
Speaker 1: car or in a clock radio. All of those devices

435
00:24:52,040 --> 00:24:55,120
Speaker 1: are pushed to talk. So those devices are never listening

436
00:24:55,480 --> 00:24:59,400
Speaker 1: unless you hit a button. If you have an Echo device,

437
00:25:00,080 --> 00:25:03,000
Speaker 1: you have the ability to hit a mute button, and

438
00:25:03,040 --> 00:25:05,119
Speaker 1: when you hit a mute button, you actually see a

439
00:25:05,160 --> 00:25:07,879
Speaker 1: red ring that goes around the outside letting you know,

440
00:25:08,400 --> 00:25:10,879
Speaker 1: and we do cut power to the microphone as well

441
00:25:11,359 --> 00:25:15,040
Speaker 1: when you hit that. Otherwise, Alexa is never listening unless

442
00:25:15,119 --> 00:25:18,320
Speaker 1: she hears her name. So when you say Alexa, then

443
00:25:18,359 --> 00:25:21,639
Speaker 1: we begin to record your voice. Your voice again is

444
00:25:21,640 --> 00:25:23,600
Speaker 1: only sent to Amazon. We do not send that to

445
00:25:23,680 --> 00:25:27,480
Speaker 1: third parties. And then you can open the Alexa app itself.

446
00:25:27,520 --> 00:25:29,880
Speaker 1: You can see every same little thing that you've ever

447
00:25:30,000 --> 00:25:33,960
Speaker 1: said to Alexa. After you've said Alexa and and then talk.

448
00:25:34,760 --> 00:25:37,040
Speaker 1: And then you also have the ability to delete any

449
00:25:37,080 --> 00:25:39,720
Speaker 1: one of those, or if you'd like, you can contact

450
00:25:39,760 --> 00:25:42,720
Speaker 1: us and remove your entire history as well. So we

451
00:25:42,760 --> 00:25:45,959
Speaker 1: do put all the control into the customer's hands. And

452
00:25:46,119 --> 00:25:49,560
Speaker 1: obviously that was this great foresight on on the part

453
00:25:49,560 --> 00:25:51,760
Speaker 1: of Amazon, because you could easily imagine that if you

454
00:25:51,760 --> 00:25:55,120
Speaker 1: did not build that into your design from the ground up,

455
00:25:55,560 --> 00:25:58,400
Speaker 1: that you would you would very quickly realize the need

456
00:25:58,520 --> 00:26:01,800
Speaker 1: for that. And if that that, that's not a good feeling.

457
00:26:02,600 --> 00:26:04,520
Speaker 1: I think in any of this, this scenario, because it

458
00:26:04,600 --> 00:26:07,600
Speaker 1: is very new for people, it's about building up trust

459
00:26:08,400 --> 00:26:11,560
Speaker 1: and it's about you know, I use terminology of crawling,

460
00:26:12,280 --> 00:26:15,560
Speaker 1: then walking, and then running, and I think as a technologist,

461
00:26:15,600 --> 00:26:18,359
Speaker 1: I always want to run, but it's important in a

462
00:26:18,400 --> 00:26:21,679
Speaker 1: space like this to start out crawling, even if that

463
00:26:21,760 --> 00:26:25,280
Speaker 1: means you're limiting what you can actually do with the device.

464
00:26:25,520 --> 00:26:28,239
Speaker 1: And one of the you know, the things that I

465
00:26:28,280 --> 00:26:31,359
Speaker 1: feel is a sign that that's been a success is

466
00:26:31,400 --> 00:26:36,160
Speaker 1: that people come and they ask for more. You know, now, hey,

467
00:26:36,200 --> 00:26:38,040
Speaker 1: I want Alexa to do more. It's okay, you know,

468
00:26:38,080 --> 00:26:40,480
Speaker 1: give me permission. I wanted to control this. I wanted

469
00:26:40,520 --> 00:26:42,400
Speaker 1: to control that, and gee, it would be great if

470
00:26:42,400 --> 00:26:45,040
Speaker 1: we call it like, so people are now they're fine

471
00:26:45,119 --> 00:26:47,640
Speaker 1: with the fact that they can talk to Alexa. Now

472
00:26:47,680 --> 00:26:49,720
Speaker 1: I want Alexa to do even more. I want Alexa

473
00:26:50,160 --> 00:26:52,520
Speaker 1: to wake up and start talking to me even if

474
00:26:52,560 --> 00:26:54,560
Speaker 1: I haven't talk to her, which is something you know

475
00:26:54,640 --> 00:26:57,560
Speaker 1: that she doesn't do today. She will never interrupt you

476
00:26:58,000 --> 00:27:02,360
Speaker 1: or start talking out of nowhere unless you first engage.

477
00:27:02,400 --> 00:27:04,560
Speaker 1: And I think that's a sign of customer trucks and

478
00:27:04,600 --> 00:27:07,280
Speaker 1: people getting excited about where the technology is headed. That's

479
00:27:07,280 --> 00:27:09,600
Speaker 1: pretty cool. I also want to say something else that

480
00:27:09,600 --> 00:27:13,240
Speaker 1: I think is really cool. Um, you guys might disagree,

481
00:27:13,600 --> 00:27:16,960
Speaker 1: and many listeners may disagree. But so I went back

482
00:27:17,000 --> 00:27:20,200
Speaker 1: and I was looking at There are blogs for specifically

483
00:27:20,240 --> 00:27:24,040
Speaker 1: for Alexa developers, which I recommend my listeners go out

484
00:27:24,160 --> 00:27:26,800
Speaker 1: check out those blogs. They are not written in a

485
00:27:26,840 --> 00:27:30,919
Speaker 1: way that is UH that's so dense or so technical

486
00:27:31,119 --> 00:27:34,040
Speaker 1: that they are inaccessible. They are very accessible and I

487
00:27:34,080 --> 00:27:36,400
Speaker 1: read over quite a few of them before we had

488
00:27:36,440 --> 00:27:39,200
Speaker 1: this conversation, and one of the reasons why I wanted

489
00:27:39,200 --> 00:27:41,960
Speaker 1: to bring this up. While they are incredibly helpful and technical,

490
00:27:42,760 --> 00:27:45,639
Speaker 1: one of the examples that was used in UH in

491
00:27:45,680 --> 00:27:49,280
Speaker 1: one of the blog posts spoke to a deep core

492
00:27:49,400 --> 00:27:52,960
Speaker 1: within me where it was actually being used to explain

493
00:27:53,040 --> 00:27:55,760
Speaker 1: what is a launch phrase? Was an invocation name and

494
00:27:55,800 --> 00:27:59,800
Speaker 1: it was about um using Dungeon dice D twenty for

495
00:28:00,320 --> 00:28:04,520
Speaker 1: that was me. Yeah, I'm as a hardcore D and

496
00:28:04,600 --> 00:28:08,040
Speaker 1: D fan from way back. I hung out with Gary

497
00:28:08,080 --> 00:28:14,200
Speaker 1: Guy Gags creator Dudgeons and Dragons. Yeah, edition campaign right now. Actually,

498
00:28:17,400 --> 00:28:21,679
Speaker 1: if I hold this up talking about the video, you

499
00:28:21,720 --> 00:28:23,679
Speaker 1: see all my D and D manuals right there. They

500
00:28:23,760 --> 00:28:26,800
Speaker 1: can my originals. I can report that Dave in fact

501
00:28:26,880 --> 00:28:30,560
Speaker 1: does have a stack approximately looks like about two ft

502
00:28:30,560 --> 00:28:34,879
Speaker 1: tall of D and D manuals behind him. Um, first edition,

503
00:28:35,000 --> 00:28:37,439
Speaker 1: all the way through a second. Oh Dave, you and

504
00:28:37,480 --> 00:28:42,240
Speaker 1: me man this blogged this room and on the phone.

505
00:28:42,680 --> 00:28:44,600
Speaker 1: Uh yeah, so that was one of those things. But

506
00:28:44,760 --> 00:28:49,000
Speaker 1: I like that the examples you guys give are interesting,

507
00:28:49,320 --> 00:28:53,320
Speaker 1: they are easy to understand, and you also it's outside

508
00:28:53,400 --> 00:28:56,800
Speaker 1: of that initial reaction I think a lot of people

509
00:28:56,840 --> 00:28:59,719
Speaker 1: would have when they hear the word Amazon. Of course,

510
00:29:00,320 --> 00:29:02,800
Speaker 1: their first thought is going to go towards online shopping

511
00:29:03,160 --> 00:29:04,720
Speaker 1: and they're thinking, oh, well, this is going to be

512
00:29:04,760 --> 00:29:06,600
Speaker 1: an app that just makes it easier for me to

513
00:29:06,600 --> 00:29:08,960
Speaker 1: buy things. But then you get into something like this

514
00:29:09,000 --> 00:29:11,080
Speaker 1: and you're like, well, no, this is here's something where

515
00:29:11,080 --> 00:29:13,640
Speaker 1: imagine you've got a table and you know, you don't

516
00:29:13,760 --> 00:29:17,480
Speaker 1: have a metric ton of dice weighing the table down.

517
00:29:17,640 --> 00:29:19,320
Speaker 1: You actually have an app and you can just call

518
00:29:19,400 --> 00:29:23,280
Speaker 1: on any time you need, and then that becomes incorporated

519
00:29:23,320 --> 00:29:26,560
Speaker 1: into your game. It's almost like Alexa is playing the

520
00:29:26,560 --> 00:29:29,880
Speaker 1: game with you, and people start to realize, oh, there's

521
00:29:30,000 --> 00:29:33,760
Speaker 1: other stuff this can do that that aren't that isn't

522
00:29:33,800 --> 00:29:36,520
Speaker 1: related to buying things. I personally think one of the

523
00:29:36,560 --> 00:29:39,640
Speaker 1: cool decisions Amazon made with this is really treating it

524
00:29:39,680 --> 00:29:43,640
Speaker 1: like an interface and as a developer, implementing that interface,

525
00:29:43,840 --> 00:29:46,720
Speaker 1: you know, uh, for your purposes is really what the

526
00:29:46,760 --> 00:29:50,040
Speaker 1: platform is is great at I mean at work. For example,

527
00:29:50,080 --> 00:29:53,040
Speaker 1: we had a hackathon recently and um, I wrote a

528
00:29:53,080 --> 00:29:55,560
Speaker 1: service in a lixer to be able to control and

529
00:29:55,600 --> 00:29:57,440
Speaker 1: this runs on a Raspberry Pie to be able to

530
00:29:57,440 --> 00:30:01,520
Speaker 1: control the servo locks on of the doors in our building.

531
00:30:01,840 --> 00:30:04,520
Speaker 1: And I recently was able to write a skill that

532
00:30:04,640 --> 00:30:08,520
Speaker 1: interacts with that Raspberry Pie via the web to open

533
00:30:08,560 --> 00:30:11,400
Speaker 1: the door, um when we see a visitor come by.

534
00:30:11,480 --> 00:30:14,880
Speaker 1: So the platform is totally open as a developer, and

535
00:30:14,880 --> 00:30:17,280
Speaker 1: and uh, there's Something else I wanted to touch on

536
00:30:17,320 --> 00:30:19,600
Speaker 1: as well, that as mentioned in the blogs and we've

537
00:30:19,640 --> 00:30:21,560
Speaker 1: kind of touched talked about a little bit, is that

538
00:30:21,640 --> 00:30:24,240
Speaker 1: the the coding side of this, it is a server side,

539
00:30:24,240 --> 00:30:28,040
Speaker 1: not a device side, not a client side kind of service. Josh,

540
00:30:28,120 --> 00:30:30,480
Speaker 1: can you talk a little bit about about why that

541
00:30:30,640 --> 00:30:32,720
Speaker 1: is just for people who might wonder like, well, why

542
00:30:32,840 --> 00:30:36,120
Speaker 1: would all this be? Why would all the the hard work,

543
00:30:36,160 --> 00:30:38,200
Speaker 1: the crunching of numbers, if you will, Why is that

544
00:30:38,280 --> 00:30:42,000
Speaker 1: happening in the cloud and not on on a dedicated device. Yeah,

545
00:30:42,000 --> 00:30:43,520
Speaker 1: that's a great question. I mean, I think there are

546
00:30:43,520 --> 00:30:46,080
Speaker 1: a couple of really solid reasons right away. You can

547
00:30:46,120 --> 00:30:48,800
Speaker 1: list for one, um, you know, the fact that the

548
00:30:48,840 --> 00:30:53,840
Speaker 1: Amazon Echo you'll likely have numerous devices, and being able

549
00:30:53,880 --> 00:30:56,040
Speaker 1: to enable that skill and have it in the cloud

550
00:30:56,280 --> 00:30:59,920
Speaker 1: um is an instantaneous thing across all of those devices,

551
00:31:00,480 --> 00:31:02,720
Speaker 1: right So, I think that's one really solid reason that

552
00:31:02,760 --> 00:31:07,160
Speaker 1: Amazon went without architecture. The other reason is moving everything

553
00:31:07,320 --> 00:31:10,440
Speaker 1: to the cloud, right keeping it on that infrastructure, it

554
00:31:10,480 --> 00:31:15,120
Speaker 1: allows for Amazon to iterate incredibly quickly on that skill

555
00:31:15,320 --> 00:31:18,480
Speaker 1: interface side of things. UM, you know, not having that

556
00:31:18,560 --> 00:31:22,640
Speaker 1: constraint to the device and worrying about shipping new hardware

557
00:31:22,680 --> 00:31:27,080
Speaker 1: to make changes over time. Amazon can simply change out

558
00:31:27,120 --> 00:31:31,520
Speaker 1: that in interface and interaction model um uh, you know

559
00:31:32,400 --> 00:31:37,160
Speaker 1: pretty much seamlessly. Um and it works um incredibly well.

560
00:31:38,200 --> 00:31:41,400
Speaker 1: And that's a that's a big part of it. It

561
00:31:41,440 --> 00:31:44,480
Speaker 1: also enables um, you know, third party developers to host

562
00:31:45,160 --> 00:31:48,120
Speaker 1: their own services wherever they want, and whatever technology they

563
00:31:48,120 --> 00:31:51,000
Speaker 1: want doesn't have to be an Amazon Cloud using any

564
00:31:51,080 --> 00:31:55,600
Speaker 1: language that they want. And we knew, you know, uh,

565
00:31:55,840 --> 00:31:58,200
Speaker 1: we thought of some pretty interesting things for Alexa, but

566
00:31:58,280 --> 00:32:02,480
Speaker 1: we knew we really need to an essence crowdsource her

567
00:32:02,520 --> 00:32:06,760
Speaker 1: ability to get smarter. Uh. And there's now over fifteen

568
00:32:06,840 --> 00:32:11,200
Speaker 1: hundred skills out there today by a third party developers

569
00:32:12,080 --> 00:32:15,160
Speaker 1: who are creating things with you know, like job described

570
00:32:15,520 --> 00:32:18,280
Speaker 1: about controlling doors and putting things in Raspberry pies and

571
00:32:18,760 --> 00:32:22,960
Speaker 1: robots and drones and health health and life sciences and

572
00:32:23,040 --> 00:32:27,360
Speaker 1: doctors offices and hospitals and games and virtual reality, you know,

573
00:32:27,480 --> 00:32:30,680
Speaker 1: things that we never would have been able to right ourselves,

574
00:32:30,680 --> 00:32:33,640
Speaker 1: that people have just taken uh and they've run with it.

575
00:32:33,720 --> 00:32:36,040
Speaker 1: And I continue to be impressed every day by what

576
00:32:36,400 --> 00:32:38,760
Speaker 1: people are doing. Yeah, exactly, Davin. I mean, you know,

577
00:32:38,880 --> 00:32:42,400
Speaker 1: you can embed you know, Alexa in if you're a

578
00:32:42,400 --> 00:32:46,440
Speaker 1: hardware designer, you can build your own version of Echo

579
00:32:46,560 --> 00:32:49,640
Speaker 1: with extended capabilities. In fact, you could have a display

580
00:32:49,960 --> 00:32:52,240
Speaker 1: or you know, if you have visions about what that

581
00:32:52,280 --> 00:32:55,480
Speaker 1: could be like, you can embed what's called Alexive Voice

582
00:32:55,480 --> 00:32:59,640
Speaker 1: services within that hardware and actually create your own version

583
00:32:59,680 --> 00:33:03,560
Speaker 1: of of the Echo, which is incredibly powerful. Interesting. So

584
00:33:04,040 --> 00:33:06,680
Speaker 1: let's talk a bit now, Josh. You you work very

585
00:33:06,720 --> 00:33:10,080
Speaker 1: closely over at Bigner branch and developing the kind of

586
00:33:10,240 --> 00:33:14,320
Speaker 1: UH curriculum that someone would who wants to develop for Alexa.

587
00:33:14,320 --> 00:33:16,400
Speaker 1: They'd find it very helpful they to to actually learn

588
00:33:16,440 --> 00:33:20,160
Speaker 1: how to code something and crafts something. And you've mentioned

589
00:33:20,160 --> 00:33:23,720
Speaker 1: before also that that one of the test skills people

590
00:33:23,720 --> 00:33:27,880
Speaker 1: will develop is one about getting information about airports and

591
00:33:28,400 --> 00:33:32,160
Speaker 1: potential delays. Let's let's kind of, in a layman sense,

592
00:33:32,280 --> 00:33:35,520
Speaker 1: kind of walk through what is the process in general

593
00:33:35,680 --> 00:33:38,680
Speaker 1: of of developing for Alexa, and then at the end

594
00:33:38,680 --> 00:33:41,560
Speaker 1: of it, I think we can do a quick little demonstration. Alexa,

595
00:33:41,640 --> 00:33:44,200
Speaker 1: by the way, I didn't introduce her. I feel like

596
00:33:44,200 --> 00:33:46,720
Speaker 1: such a cad I didn't introduce her, but she is

597
00:33:46,800 --> 00:33:49,800
Speaker 1: also in the studio with us. So Alexa, how are

598
00:33:49,800 --> 00:33:54,520
Speaker 1: you doing today? Great. And if you're listening, make sure

599
00:33:54,560 --> 00:33:58,280
Speaker 1: you're muted. Although you were already listening your your echo

600
00:33:58,320 --> 00:34:01,160
Speaker 1: has probably gone on several times. My my parents are

601
00:34:01,200 --> 00:34:04,040
Speaker 1: going to send me a very mad email. My parents,

602
00:34:04,080 --> 00:34:07,240
Speaker 1: by the way, they have they're the ones who asked

603
00:34:07,240 --> 00:34:10,239
Speaker 1: me to do this podcast because my parents own own

604
00:34:10,320 --> 00:34:12,680
Speaker 1: and Echo, and they talked to it all the time,

605
00:34:12,719 --> 00:34:15,040
Speaker 1: and they demonstrated to me all the time. And uh,

606
00:34:15,080 --> 00:34:17,640
Speaker 1: the only reason I haven't picked one up yet is

607
00:34:17,680 --> 00:34:19,120
Speaker 1: I've got to wait for my wife to go out

608
00:34:19,120 --> 00:34:21,319
Speaker 1: of town just long enough so I can get it

609
00:34:21,320 --> 00:34:23,560
Speaker 1: and incorporated into everything, so that when she comes back

610
00:34:23,560 --> 00:34:26,600
Speaker 1: and sees that I bought it, it's so awesome and

611
00:34:26,640 --> 00:34:29,080
Speaker 1: incorporated and integrated that she would never want to get

612
00:34:29,160 --> 00:34:32,640
Speaker 1: rid of it. Yeah, I run the white I do

613
00:34:32,719 --> 00:34:38,040
Speaker 1: the wife test with technology as well, and is slowly

614
00:34:38,120 --> 00:34:42,480
Speaker 1: warmed up to She was resisted at first, but she's

615
00:34:42,560 --> 00:34:45,759
Speaker 1: just accepted it into our home. Step one is is

616
00:34:45,800 --> 00:34:48,759
Speaker 1: get my wife comfortable with Alexa. Step two is to

617
00:34:48,760 --> 00:34:51,920
Speaker 1: get her into a D and D campaign. All right,

618
00:34:52,000 --> 00:34:54,799
Speaker 1: So so we're I come to you, Josh, I just say,

619
00:34:54,840 --> 00:34:57,320
Speaker 1: I say I am interested in starting to develop for Alexa.

620
00:34:57,719 --> 00:35:01,440
Speaker 1: What's kind of the process of of learning to develop

621
00:35:01,480 --> 00:35:05,160
Speaker 1: and and the process of actually developing in a skill. Yeah.

622
00:35:05,239 --> 00:35:08,120
Speaker 1: And so you mentioned earlier that you had read our

623
00:35:08,160 --> 00:35:11,480
Speaker 1: blog and you followed that arc. It's um as you

624
00:35:11,520 --> 00:35:14,320
Speaker 1: get into the process, it's really a couple of things

625
00:35:14,400 --> 00:35:17,680
Speaker 1: that you get your brain wrapped around. Initially, you know,

626
00:35:17,800 --> 00:35:20,280
Speaker 1: most developers are going to be coming from a graphical

627
00:35:20,400 --> 00:35:23,560
Speaker 1: user interface background, right, I mean that's the lay of

628
00:35:23,600 --> 00:35:27,040
Speaker 1: the land these days. Of the applications being written, though

629
00:35:27,040 --> 00:35:30,399
Speaker 1: that's changing, are all graphical, right, because it's either meant

630
00:35:30,440 --> 00:35:36,040
Speaker 1: for a smartphone screen or it's on a laptop or desktop. Yeah. Once, once,

631
00:35:36,080 --> 00:35:41,800
Speaker 1: once mac os and and and Microsoft Windows really took hold,

632
00:35:42,200 --> 00:35:45,960
Speaker 1: we started to see other types of UIs disappear, and

633
00:35:46,000 --> 00:35:48,960
Speaker 1: then we just began to assume that was the only

634
00:35:48,960 --> 00:35:52,439
Speaker 1: way you could do things for a while. Yeah, exactly right. Uh.

635
00:35:52,520 --> 00:35:55,080
Speaker 1: You know, however, the paradigm as we see is changing.

636
00:35:55,960 --> 00:36:00,880
Speaker 1: So the first thing to understand is the skill interface

637
00:36:01,000 --> 00:36:04,200
Speaker 1: portion of what I consider to be a two part thing.

638
00:36:04,239 --> 00:36:07,680
Speaker 1: It's a skill interface and then a skill service. The

639
00:36:07,719 --> 00:36:11,560
Speaker 1: skill interface, like I said, is responsible for resolving the

640
00:36:11,640 --> 00:36:15,000
Speaker 1: user's words to the first step of building a skill,

641
00:36:15,040 --> 00:36:18,840
Speaker 1: which is defining the interaction model, and so I consider

642
00:36:19,200 --> 00:36:22,400
Speaker 1: consider that to be setting an invocation name, which we

643
00:36:22,440 --> 00:36:25,759
Speaker 1: talked about earlier. It's the name of your skill, um

644
00:36:25,760 --> 00:36:28,120
Speaker 1: and how a user is going to be communicating with it.

645
00:36:28,239 --> 00:36:30,319
Speaker 1: So this is so that you can you can create

646
00:36:30,360 --> 00:36:33,040
Speaker 1: a distinction between the skill that you're developing in every

647
00:36:33,040 --> 00:36:36,600
Speaker 1: other skill that's out there on available on with Alexa. Yeah,

648
00:36:36,600 --> 00:36:38,400
Speaker 1: that's exactly right, and it's going to be unique to

649
00:36:38,400 --> 00:36:41,080
Speaker 1: the skill that you're building. Um. Amazon will in fact,

650
00:36:41,400 --> 00:36:43,320
Speaker 1: uh not allow you to have you know, of course,

651
00:36:43,400 --> 00:36:47,319
Speaker 1: the same name of another skill that that introduces an

652
00:36:47,320 --> 00:36:51,200
Speaker 1: interesting new like domain squatting type problem, that doesn't wonder,

653
00:36:51,280 --> 00:36:54,160
Speaker 1: I wonder what will become of that at any rate? Um.

654
00:36:54,239 --> 00:36:57,319
Speaker 1: So in our class, for example, one skill that we

655
00:36:57,360 --> 00:36:59,560
Speaker 1: teach you to write, which gives you a pretty good

656
00:36:59,560 --> 00:37:03,200
Speaker 1: cross action of all the different capabilities of the platform

657
00:37:03,320 --> 00:37:06,399
Speaker 1: is the airport info skill. So my first step would

658
00:37:06,400 --> 00:37:10,799
Speaker 1: be defining the invocation name of airport info in Amazon's

659
00:37:10,880 --> 00:37:14,520
Speaker 1: skill interface. And this is literally a web portal that

660
00:37:14,560 --> 00:37:18,680
Speaker 1: you visit and configure in your web browser. So as

661
00:37:18,719 --> 00:37:20,520
Speaker 1: you can figure that. The next step that you get

662
00:37:20,560 --> 00:37:25,080
Speaker 1: to is defining the sample utterances that as a developer,

663
00:37:25,560 --> 00:37:29,080
Speaker 1: I need to hook into UM as a result of

664
00:37:29,120 --> 00:37:32,960
Speaker 1: the resolution of what the interaction model said someone had

665
00:37:33,000 --> 00:37:37,440
Speaker 1: just spoken into the device, right. So this is the

666
00:37:37,560 --> 00:37:39,920
Speaker 1: kind of magic that we're talking about earlier. The black

667
00:37:40,000 --> 00:37:44,440
Speaker 1: box that Amazon offers you UM that leverages the artificial

668
00:37:44,480 --> 00:37:48,000
Speaker 1: intelligence and machine learning, you know, the cutting edge technology

669
00:37:48,160 --> 00:37:52,600
Speaker 1: really there, UM, and we train that model up with

670
00:37:52,800 --> 00:37:58,000
Speaker 1: sample utterances. So those sample utterances for airport information UM.

671
00:37:58,040 --> 00:38:01,040
Speaker 1: It's actually UH, it's got a couple of different aspects.

672
00:38:01,160 --> 00:38:04,880
Speaker 1: One is, UM, we want to be able to resolve

673
00:38:05,280 --> 00:38:08,560
Speaker 1: the various ways that someone will ask for information about

674
00:38:08,560 --> 00:38:11,520
Speaker 1: an airport. And then secondly, we've got to have a

675
00:38:11,560 --> 00:38:15,480
Speaker 1: way to pass a variable into the application or the skill.

676
00:38:15,680 --> 00:38:18,560
Speaker 1: So I want to be able to, for example, determine

677
00:38:18,880 --> 00:38:22,719
Speaker 1: if someone had said SFO or a t l UM. Right,

678
00:38:22,800 --> 00:38:25,640
Speaker 1: they're asking about it a specific airport code, and I

679
00:38:25,680 --> 00:38:27,600
Speaker 1: need to be able to throw that into a variable.

680
00:38:28,040 --> 00:38:31,080
Speaker 1: So the interaction model has a mechanism for doing that

681
00:38:31,120 --> 00:38:35,479
Speaker 1: called a slot, and a slot is it's kind of UM.

682
00:38:35,520 --> 00:38:38,880
Speaker 1: It's Amazon's terminology for basically a variable assignment that we

683
00:38:38,880 --> 00:38:41,719
Speaker 1: want to be able to do. So in that in

684
00:38:41,760 --> 00:38:46,480
Speaker 1: that set of sample utterances, we give strings that represent

685
00:38:46,640 --> 00:38:49,640
Speaker 1: how user could ask for airport information. So I might

686
00:38:49,680 --> 00:38:52,680
Speaker 1: ask for, uh, you know, Alexa, ask airport info for

687
00:38:52,840 --> 00:38:56,640
Speaker 1: flight delays at a t L, um for flight information

688
00:38:56,719 --> 00:38:59,320
Speaker 1: at a t L, for delay status at a t L.

689
00:38:59,440 --> 00:39:01,560
Speaker 1: You see the very creations of ways that someone could

690
00:39:01,560 --> 00:39:05,319
Speaker 1: ask for that info. UM. Now, like I said, it's generalized,

691
00:39:05,400 --> 00:39:08,720
Speaker 1: so in other words, it's a fuzzy resolution between those

692
00:39:08,719 --> 00:39:13,319
Speaker 1: phrases and UM. An intent. So an intent is an

693
00:39:13,320 --> 00:39:15,960
Speaker 1: indication of what someone would like your skill to do.

694
00:39:17,000 --> 00:39:18,839
Speaker 1: And we want our skill to be able to give

695
00:39:18,920 --> 00:39:22,760
Speaker 1: us information about an airport. And so the skill interface,

696
00:39:22,840 --> 00:39:25,439
Speaker 1: once we've set up that sample utterances list and we've

697
00:39:25,480 --> 00:39:29,760
Speaker 1: said okay, the words that fall at this portion of

698
00:39:29,960 --> 00:39:32,560
Speaker 1: what someone says are actually going to be dropped into

699
00:39:32,600 --> 00:39:35,680
Speaker 1: this new thing called a slot. And I want to

700
00:39:35,680 --> 00:39:38,040
Speaker 1: be able to have that as a variable on the

701
00:39:38,080 --> 00:39:41,360
Speaker 1: second portion of building a skill, which is the skill service.

702
00:39:41,960 --> 00:39:44,279
Speaker 1: So once the skill interface has done that work for me,

703
00:39:44,680 --> 00:39:47,360
Speaker 1: and since the information of what it found of what

704
00:39:47,480 --> 00:39:50,160
Speaker 1: someone had said to the skill service, that's where we're

705
00:39:50,160 --> 00:39:53,840
Speaker 1: in node land, or like David said, it's really any

706
00:39:54,040 --> 00:39:57,560
Speaker 1: programming language that can speak HTTPS and live on a

707
00:39:57,600 --> 00:40:00,719
Speaker 1: server One of the steps that actually happened at that

708
00:40:00,760 --> 00:40:04,720
Speaker 1: point too, is after you've set up everything that johsh

709
00:40:04,719 --> 00:40:07,799
Speaker 1: talked about, in essence, this is what I want to

710
00:40:07,800 --> 00:40:11,160
Speaker 1: talk about. Here's some examples of me using it in

711
00:40:11,160 --> 00:40:13,920
Speaker 1: a sentence. That's where the computer science comes in. So

712
00:40:14,000 --> 00:40:17,280
Speaker 1: that's where it runs in the cloud, and we actually

713
00:40:17,280 --> 00:40:21,480
Speaker 1: do a bunch of uh, you know, AI and machine

714
00:40:21,560 --> 00:40:24,240
Speaker 1: learning and everything that sits on top of Amazon Web Services,

715
00:40:24,280 --> 00:40:27,920
Speaker 1: and we generate a lexicon and a language model, so

716
00:40:28,360 --> 00:40:31,560
Speaker 1: it's it's all done ahead of time, and it's almost

717
00:40:31,560 --> 00:40:34,160
Speaker 1: a you know, me being a sci fi geek, it's

718
00:40:34,160 --> 00:40:37,440
Speaker 1: almost like the matrix when Neo says, teach me kung fu,

719
00:40:37,800 --> 00:40:39,919
Speaker 1: and then he's like, I now know kung fu. That's

720
00:40:39,960 --> 00:40:42,960
Speaker 1: the part that we do through the portal. So now

721
00:40:43,000 --> 00:40:46,520
Speaker 1: Alexa knows kong fu, uh, and she can talk to

722
00:40:46,560 --> 00:40:48,600
Speaker 1: you about kong fu. But you actually have to build

723
00:40:48,640 --> 00:40:51,520
Speaker 1: something to be able to respond to that. And that's

724
00:40:51,560 --> 00:40:53,640
Speaker 1: what Josh is talking about now when we move on

725
00:40:53,680 --> 00:40:56,160
Speaker 1: to the node piece, right, Yeah, so this is the

726
00:40:56,200 --> 00:40:59,279
Speaker 1: magic black box in the cloud that has you know,

727
00:40:59,360 --> 00:41:03,239
Speaker 1: the cutting edge artificial intelligence technology that Amazon offers us

728
00:41:03,960 --> 00:41:07,080
Speaker 1: um on the interaction model side of things, and as

729
00:41:07,080 --> 00:41:11,080
Speaker 1: a developer, I provide a training data. Uh. When David

730
00:41:11,120 --> 00:41:14,160
Speaker 1: said it's sort of prepares things ahead of time. It

731
00:41:14,200 --> 00:41:19,160
Speaker 1: literally bakes that training data down into something that occurs,

732
00:41:19,280 --> 00:41:22,279
Speaker 1: you know, in near real time. That interaction is is

733
00:41:22,400 --> 00:41:25,760
Speaker 1: very very fast and there's there's little to no latency

734
00:41:26,040 --> 00:41:28,640
Speaker 1: um and the reason is because they use that sample

735
00:41:28,680 --> 00:41:31,839
Speaker 1: data to prepare a model ahead of time. Uh, it's

736
00:41:31,880 --> 00:41:35,080
Speaker 1: really cool. It's it's super cool. It's you know, we're

737
00:41:35,160 --> 00:41:37,520
Speaker 1: we're living a sci fi novel right now. I feel

738
00:41:37,520 --> 00:41:39,759
Speaker 1: like I'm in cryptomic con or. So well, it's it's

739
00:41:39,760 --> 00:41:42,279
Speaker 1: got to be exciting too to be a developer that

740
00:41:42,400 --> 00:41:47,440
Speaker 1: gets to take advantage of that level of of technology

741
00:41:47,760 --> 00:41:50,960
Speaker 1: and not have to build it yourself, Like to to

742
00:41:51,360 --> 00:41:54,279
Speaker 1: develop on top of something that has already got this

743
00:41:54,320 --> 00:41:57,080
Speaker 1: amazing capability and on the shoulders of giants. I mean,

744
00:41:57,080 --> 00:42:00,279
Speaker 1: we're leveraging all of that platform and infrastry sure and

745
00:42:00,400 --> 00:42:03,799
Speaker 1: research that's that's occurred there. Uh. And right now there

746
00:42:03,880 --> 00:42:06,719
Speaker 1: is a language model out in the cloud that knows

747
00:42:06,800 --> 00:42:10,040
Speaker 1: all about rolling dungeons and dragon dice, which is It's

748
00:42:10,040 --> 00:42:12,520
Speaker 1: a reassuring thought, isn't it. I would say that that

749
00:42:12,600 --> 00:42:16,239
Speaker 1: was critical and critical hit it too a good initiative

750
00:42:16,680 --> 00:42:21,600
Speaker 1: in doing that, I'm terrible. I've lost my saving throw.

751
00:42:21,640 --> 00:42:27,279
Speaker 1: It gets being cool, that's not she will actually she'll

752
00:42:27,280 --> 00:42:31,400
Speaker 1: make a little scathing comment about you rolling a one.

753
00:42:31,880 --> 00:42:33,759
Speaker 1: That's that's pa why I showed it so like if

754
00:42:33,800 --> 00:42:36,400
Speaker 1: you could aller it. But the big light of my

755
00:42:36,480 --> 00:42:39,400
Speaker 1: kids rolling some dice if they had one, and she'll say, hey,

756
00:42:39,440 --> 00:42:41,520
Speaker 1: do you have a warple sword? If you hit twenty,

757
00:42:41,800 --> 00:42:46,120
Speaker 1: oh nice because off the head there you go. Uh so,

758
00:42:46,480 --> 00:42:48,399
Speaker 1: all right, So you you get to this point where

759
00:42:48,440 --> 00:42:51,440
Speaker 1: you've you've defined what it is you you know, incode,

760
00:42:51,440 --> 00:42:53,799
Speaker 1: You've defined what it is that the action that needs

761
00:42:53,840 --> 00:42:56,799
Speaker 1: to happen, um and you've you've built it. So we've

762
00:42:56,800 --> 00:42:59,360
Speaker 1: provided the training data and the interaction model, right, and

763
00:42:59,400 --> 00:43:04,200
Speaker 1: we've said when you uh hear words that are along

764
00:43:04,239 --> 00:43:08,560
Speaker 1: the lines of our request for airport information, I want

765
00:43:08,600 --> 00:43:11,960
Speaker 1: you to send an airport info intent as part of

766
00:43:11,960 --> 00:43:15,239
Speaker 1: a Jason payload to my skill service. And I also

767
00:43:15,280 --> 00:43:17,640
Speaker 1: want you to grab I want you to wrap up

768
00:43:17,719 --> 00:43:21,000
Speaker 1: what they said at this portion of that utterance into

769
00:43:21,040 --> 00:43:24,600
Speaker 1: a variable that I can use on my skill service. UM.

770
00:43:24,640 --> 00:43:28,120
Speaker 1: So the skill service in our class is living on UM.

771
00:43:28,160 --> 00:43:32,600
Speaker 1: It's actually another Amazon service called Lambda UM, which is

772
00:43:32,640 --> 00:43:35,880
Speaker 1: an HTPS server that spins up and shuts down. It's

773
00:43:35,920 --> 00:43:39,440
Speaker 1: kind of like Hiroku. If anyone's ever done rubyond rails

774
00:43:39,480 --> 00:43:42,280
Speaker 1: development here, Hiroku is a real go to for handling

775
00:43:42,280 --> 00:43:44,799
Speaker 1: the DevOps side of things UM and it's kind of

776
00:43:44,800 --> 00:43:47,560
Speaker 1: an on demand server platform that we're using in the

777
00:43:47,560 --> 00:43:53,200
Speaker 1: class UH and so on that. On that AWS Lambda instance,

778
00:43:53,280 --> 00:43:58,160
Speaker 1: we've written node code that handles parsing the HTTPS UM

779
00:43:58,320 --> 00:44:01,239
Speaker 1: Jason payload that comes across from the Skill interface and

780
00:44:01,280 --> 00:44:04,400
Speaker 1: it says, Okay, I've gotten Jason information here, and I

781
00:44:04,440 --> 00:44:09,400
Speaker 1: received an airport info intent UM, and I've got a

782
00:44:09,480 --> 00:44:12,920
Speaker 1: variable of a t L here, And with that information

783
00:44:13,239 --> 00:44:15,200
Speaker 1: I can then do what I will with it. I

784
00:44:15,200 --> 00:44:17,360
Speaker 1: mean I can I can go make a web request

785
00:44:17,640 --> 00:44:20,640
Speaker 1: or right to a database I could call off to

786
00:44:20,680 --> 00:44:23,480
Speaker 1: another service UM. And that's actually what we do with

787
00:44:23,520 --> 00:44:27,680
Speaker 1: Airport info is we hit the Federal Aviation Administration's servers

788
00:44:28,320 --> 00:44:31,400
Speaker 1: and request the status for a t L. Then we

789
00:44:31,480 --> 00:44:34,440
Speaker 1: build a string which is what Alexa is going to

790
00:44:34,480 --> 00:44:37,760
Speaker 1: respond with, and send it back to the Skill interface

791
00:44:38,680 --> 00:44:41,600
Speaker 1: and that gets forwarded onto the device. So that's kind

792
00:44:41,600 --> 00:44:44,320
Speaker 1: of the round trip of how an interaction with the

793
00:44:44,360 --> 00:44:46,839
Speaker 1: skill would work and what we would do as the developer. There,

794
00:44:46,960 --> 00:44:50,200
Speaker 1: so got the skill interface, skill service, skill service response

795
00:44:50,280 --> 00:44:52,120
Speaker 1: to the interface and hands it back to the device.

796
00:44:52,239 --> 00:44:58,040
Speaker 1: Now for practical experience for the person who's actually using Alexa. Obviously,

797
00:44:58,440 --> 00:45:01,200
Speaker 1: all of that information, while awesome, is not something that

798
00:45:01,200 --> 00:45:02,960
Speaker 1: you have to worry about. If you did have to

799
00:45:03,000 --> 00:45:06,040
Speaker 1: worry about it, then you would not really have a

800
00:45:06,800 --> 00:45:09,560
Speaker 1: consumer product in your hands. You'd be a developer. But

801
00:45:10,080 --> 00:45:12,880
Speaker 1: in order for you to understand what is what's the

802
00:45:12,960 --> 00:45:15,239
Speaker 1: end result with that, I thought i'd be cool, Josh,

803
00:45:15,320 --> 00:45:18,400
Speaker 1: would you mind asking Alexa kind to give an example

804
00:45:18,480 --> 00:45:21,440
Speaker 1: of that app in action? Yeah? Sure, So it's skill

805
00:45:21,480 --> 00:45:23,960
Speaker 1: I should say I have. Don't worry about it. That's

806
00:45:23,960 --> 00:45:26,759
Speaker 1: a common blunder initially coming into the thing. I think

807
00:45:26,800 --> 00:45:30,120
Speaker 1: I said app once today as well, So we don't

808
00:45:30,160 --> 00:45:34,680
Speaker 1: have any app. That's fair, that's fair, that's a good

809
00:45:34,719 --> 00:45:39,279
Speaker 1: skill idea. Yeah, so alright, so let me first, let's

810
00:45:39,360 --> 00:45:44,120
Speaker 1: un mute her, all right? So okay, So Alexa ask

811
00:45:44,200 --> 00:45:48,200
Speaker 1: airport info for flight delays at a t L. There

812
00:45:48,280 --> 00:45:51,759
Speaker 1: is currently no delay at Hartsfield Jets in Atlanta International

813
00:45:53,080 --> 00:45:56,759
Speaker 1: ladies and gentlemen, This is an amazing day. Not only

814
00:45:56,760 --> 00:46:01,280
Speaker 1: did we get to hear voice wrecking shouldn't service work

815
00:46:01,600 --> 00:46:03,520
Speaker 1: in real time on this show? But there are no

816
00:46:03,640 --> 00:46:11,640
Speaker 1: flight delays at the airport. Actually yeah, it's it's it's

817
00:46:11,640 --> 00:46:13,600
Speaker 1: not calling it's not really calling the airport, it's not

818
00:46:13,640 --> 00:46:16,560
Speaker 1: calling the f A uh. But this is this is

819
00:46:16,600 --> 00:46:21,320
Speaker 1: really interesting again showing just a very simple application. Obviously,

820
00:46:21,360 --> 00:46:25,560
Speaker 1: you could ask for lots of different things. Sorry, skill,

821
00:46:26,480 --> 00:46:28,600
Speaker 1: but you could ask for lots of different things, including

822
00:46:28,680 --> 00:46:31,080
Speaker 1: like if you were curious about what's the weather going

823
00:46:31,120 --> 00:46:33,160
Speaker 1: to be? Like what what was the time of day?

824
00:46:33,760 --> 00:46:36,640
Speaker 1: Can you play my favorite playlist on such and such?

825
00:46:37,040 --> 00:46:40,000
Speaker 1: All of these sort of things that are all really

826
00:46:40,040 --> 00:46:41,680
Speaker 1: just you know, if you come up with an idea

827
00:46:41,760 --> 00:46:45,120
Speaker 1: that could be uh essentially if you could if you

828
00:46:45,120 --> 00:46:49,120
Speaker 1: could do it across a computer web browser, and if

829
00:46:49,160 --> 00:46:52,800
Speaker 1: you're able to translate that into an experience that works,

830
00:46:53,480 --> 00:46:58,759
Speaker 1: especially speech um in a speech role, then it's it's

831
00:46:58,920 --> 00:47:01,960
Speaker 1: totally possible. And only that, but I saw because I

832
00:47:02,000 --> 00:47:04,239
Speaker 1: was curious about this. There are some things that we

833
00:47:04,320 --> 00:47:08,000
Speaker 1: do with computers where if you were asking a question,

834
00:47:08,480 --> 00:47:11,160
Speaker 1: you might need a little more information or a little

835
00:47:11,200 --> 00:47:14,040
Speaker 1: bit more of Uh, it might be a little difficult

836
00:47:14,280 --> 00:47:18,200
Speaker 1: to explain something just simply in spoken word. Uh. I

837
00:47:18,280 --> 00:47:21,360
Speaker 1: find that all the time doing an audio podcast, that

838
00:47:21,480 --> 00:47:25,240
Speaker 1: it can be kind of challenging to explain a certain

839
00:47:25,280 --> 00:47:28,200
Speaker 1: concept without the use of visual aids. I saw that

840
00:47:28,320 --> 00:47:30,799
Speaker 1: with Alexa you could actually pair that with something like

841
00:47:30,840 --> 00:47:33,399
Speaker 1: a display on an app where you can you can

842
00:47:33,440 --> 00:47:36,480
Speaker 1: have a little more information about whatever the request is. Yeah,

843
00:47:36,480 --> 00:47:40,919
Speaker 1: that's right. Yes, So Amazon also provides a additional part

844
00:47:40,960 --> 00:47:44,479
Speaker 1: of that response from the server called a card um.

845
00:47:44,520 --> 00:47:47,280
Speaker 1: So if we sent that information onto the skill interface,

846
00:47:47,400 --> 00:47:51,440
Speaker 1: what would happen is we could preserve or send additional

847
00:47:51,480 --> 00:47:56,040
Speaker 1: info to the Alexa companion app, which is in your browser,

848
00:47:56,080 --> 00:47:59,200
Speaker 1: it's on your phone, it's on the Android and Iosh

849
00:47:59,400 --> 00:48:02,280
Speaker 1: client the interfaces with it, and we'd have a history

850
00:48:02,560 --> 00:48:05,439
Speaker 1: um and additional info as well um that we could

851
00:48:05,440 --> 00:48:08,839
Speaker 1: display with that card mechanism. Right, that's a great way

852
00:48:08,840 --> 00:48:11,839
Speaker 1: of getting around what could be a real challenge. I mean,

853
00:48:11,880 --> 00:48:15,719
Speaker 1: that's something that because we've designed so much of our

854
00:48:15,760 --> 00:48:21,800
Speaker 1: interaction to be primarily a visual experience, there's certain tasks

855
00:48:21,840 --> 00:48:25,200
Speaker 1: that you do that don't translate as easily into something

856
00:48:25,239 --> 00:48:28,200
Speaker 1: that's audio based. Yeah, it's a really interesting aspect of

857
00:48:28,280 --> 00:48:30,839
Speaker 1: voice user interface design as well. I mean it's such

858
00:48:30,880 --> 00:48:35,600
Speaker 1: a um ephemeral format. It's it's here and then it's gone,

859
00:48:35,960 --> 00:48:39,520
Speaker 1: and it's an interesting problem coming from you know, the

860
00:48:39,520 --> 00:48:43,719
Speaker 1: graphical user interface background that everybody's got UM. I think

861
00:48:43,719 --> 00:48:46,200
Speaker 1: being able to hook a little bit into the g

862
00:48:46,400 --> 00:48:49,120
Speaker 1: u I side of things and persist data UM and

863
00:48:49,160 --> 00:48:51,400
Speaker 1: allow people to refer to that back um. You know,

864
00:48:51,520 --> 00:48:55,719
Speaker 1: it does give your skill legs beyond just that one interaction.

865
00:48:56,080 --> 00:49:00,320
Speaker 1: You know, for situational experiences, an on demand info voice

866
00:49:00,320 --> 00:49:03,040
Speaker 1: is great, but being able to carry it beyond that

867
00:49:03,200 --> 00:49:05,280
Speaker 1: is also one of the things that the platform offers.

868
00:49:05,640 --> 00:49:08,640
Speaker 1: That's fantastic. Now I have a question for both of you.

869
00:49:08,680 --> 00:49:10,920
Speaker 1: I'm going to kind of start wrapping this up because

870
00:49:11,520 --> 00:49:14,080
Speaker 1: I feel like we've got a nice, a nice foundation

871
00:49:14,160 --> 00:49:16,399
Speaker 1: for a discussion here and I think that it will

872
00:49:16,480 --> 00:49:19,960
Speaker 1: really uh, it's really enlightening to understand, like what are

873
00:49:20,000 --> 00:49:22,680
Speaker 1: the challenges not just the challenges, but what are the

874
00:49:23,120 --> 00:49:26,600
Speaker 1: potential benefits of this kind of technology. My question to

875
00:49:26,680 --> 00:49:29,480
Speaker 1: both of you, and it's a pretty simple one, is

876
00:49:30,239 --> 00:49:33,520
Speaker 1: what is your personal favorite skill that you've had a

877
00:49:33,600 --> 00:49:36,239
Speaker 1: chance to play with on Alexa and Dave, I'm gonna

878
00:49:36,280 --> 00:49:40,799
Speaker 1: have you go first. Yeah. So, um, besides, you know,

879
00:49:41,200 --> 00:49:44,279
Speaker 1: using the dice with my kids the funniest one, and

880
00:49:44,320 --> 00:49:46,560
Speaker 1: this is this speaks I think a little bit to

881
00:49:46,800 --> 00:49:50,839
Speaker 1: my maturity level. Is I have this skill enabled. It's

882
00:49:50,880 --> 00:49:54,480
Speaker 1: called for a fart, and you can ask for a

883
00:49:54,600 --> 00:49:58,080
Speaker 1: fart and you will get one through Alexa. So I

884
00:49:58,200 --> 00:50:02,880
Speaker 1: actually I have dots so upstairs in my house. Yeah,

885
00:50:02,920 --> 00:50:05,839
Speaker 1: all right, there's a dot in each one of my

886
00:50:05,920 --> 00:50:08,880
Speaker 1: kid's rooms. And then I have a full echo in

887
00:50:08,920 --> 00:50:11,600
Speaker 1: my bedroom. And there's actually a place in that upper

888
00:50:11,640 --> 00:50:14,600
Speaker 1: hallway that I can say Alexa asked for a fart,

889
00:50:14,680 --> 00:50:20,200
Speaker 1: and there is a symphony musical returned across the entire

890
00:50:20,280 --> 00:50:24,959
Speaker 1: upstairs and I immediately here, Dad, not again. A nice,

891
00:50:25,040 --> 00:50:30,160
Speaker 1: a nice chorus followed by groans. I can appreciate that

892
00:50:30,200 --> 00:50:35,520
Speaker 1: as someone who loves puns and and grown and grown worthy. Hum. Um, yeah,

893
00:50:35,560 --> 00:50:39,239
Speaker 1: that's fantastic. All right, So Josh, you your answer, that's

894
00:50:39,239 --> 00:50:42,400
Speaker 1: an interesting one. Well, I have to say on the

895
00:50:42,440 --> 00:50:46,480
Speaker 1: pragmatic side of things, Um, I have used lift skill

896
00:50:47,080 --> 00:50:49,919
Speaker 1: to great effect. So you just say, Alexa asked lift

897
00:50:49,960 --> 00:50:52,000
Speaker 1: for a ride, and guess what a car shows up

898
00:50:52,560 --> 00:50:55,759
Speaker 1: at your door right to pick you up. Amazing, and

899
00:50:55,800 --> 00:50:57,799
Speaker 1: it's such a good fit for the platform as well.

900
00:50:58,160 --> 00:51:00,040
Speaker 1: I mean, you're walking out the door, you say, I

901
00:51:00,080 --> 00:51:01,920
Speaker 1: don't want to get my phone out. Let's let's have

902
00:51:01,960 --> 00:51:05,040
Speaker 1: a car show up right now. Uh, and it happens.

903
00:51:05,120 --> 00:51:09,640
Speaker 1: That's very convenient. And there is a if you you know,

904
00:51:09,640 --> 00:51:11,120
Speaker 1: we talked a little bit about D and D. If

905
00:51:11,160 --> 00:51:13,839
Speaker 1: you ever played um that goes all the way back

906
00:51:13,880 --> 00:51:17,120
Speaker 1: into the old bulletin board systems days of Space Empire

907
00:51:17,480 --> 00:51:20,959
Speaker 1: or anything like that, there is a skill called star

908
00:51:21,120 --> 00:51:24,880
Speaker 1: Lanes by Joe Jo Quinta that basically allows you to

909
00:51:25,040 --> 00:51:27,440
Speaker 1: do that and you can play It's a multi player

910
00:51:27,560 --> 00:51:31,799
Speaker 1: online Space Empire game through Alexa, and that is one

911
00:51:31,840 --> 00:51:34,040
Speaker 1: of my other favorites. I would highly recommend checking that

912
00:51:34,080 --> 00:51:36,480
Speaker 1: out if you've ever played any of those games in

913
00:51:36,480 --> 00:51:38,760
Speaker 1: the past. That could be quite addicting. That's really cool,

914
00:51:38,920 --> 00:51:42,600
Speaker 1: I mean, and it really does speak to the potential

915
00:51:42,719 --> 00:51:47,160
Speaker 1: for things that we can't even necessarily anticipate right now

916
00:51:47,400 --> 00:51:50,440
Speaker 1: that could end up being either uh, it could end

917
00:51:50,520 --> 00:51:52,880
Speaker 1: up being something where you know, people talk about for

918
00:51:52,880 --> 00:51:54,919
Speaker 1: a little while like, oh, that's really clever. It's really

919
00:51:54,920 --> 00:51:57,400
Speaker 1: neat use of the technology, or it could truly be

920
00:51:57,480 --> 00:52:01,239
Speaker 1: transformative to the point where we didn't think about this

921
00:52:01,360 --> 00:52:03,480
Speaker 1: and now we can't think of what what life would

922
00:52:03,480 --> 00:52:05,840
Speaker 1: be without it. I mean, that's the that's the cool

923
00:52:06,360 --> 00:52:09,240
Speaker 1: promise of this kind of tech is that it stands

924
00:52:09,280 --> 00:52:13,880
Speaker 1: to be really disruptive, uh, for a type of technology

925
00:52:13,920 --> 00:52:17,760
Speaker 1: that's been fairly set in its ways for the last

926
00:52:17,840 --> 00:52:21,359
Speaker 1: several decades. And um, I love seeing this. I love

927
00:52:21,360 --> 00:52:24,239
Speaker 1: the discussions about machine learning and artificial intelligence. I love

928
00:52:24,320 --> 00:52:27,560
Speaker 1: the discussions about natural language and the challenges that we

929
00:52:27,600 --> 00:52:30,360
Speaker 1: face when we try to create interfaces that can accept

930
00:52:30,440 --> 00:52:33,560
Speaker 1: natural language as an input. Guys, I have to thank

931
00:52:33,560 --> 00:52:36,520
Speaker 1: you so much for joining me on tech Stuff, Dave,

932
00:52:36,719 --> 00:52:39,880
Speaker 1: Thank you, Josh, thank you. I really hope that you

933
00:52:40,000 --> 00:52:43,399
Speaker 1: enjoyed your time here on on David. I know you're

934
00:52:43,400 --> 00:52:47,400
Speaker 1: not actually in the studio, but I'll pick up my

935
00:52:47,560 --> 00:52:50,560
Speaker 1: laptop and I'll just show you around in a minute.

936
00:52:50,600 --> 00:52:53,799
Speaker 1: Thank you. Yeah. I really enjoyed being here today and

937
00:52:53,840 --> 00:52:56,000
Speaker 1: having this conversation. Thank you so much for having me

938
00:52:56,040 --> 00:52:58,160
Speaker 1: on the show. Absolutely yeah, thanks for having us, Jonathan,

939
00:52:58,200 --> 00:53:02,000
Speaker 1: I was really stimulating conversation. Uh. Yeah. And Guys, like

940
00:53:02,040 --> 00:53:04,960
Speaker 1: I said, if you want to learn more about developing

941
00:53:05,000 --> 00:53:06,960
Speaker 1: for Alexa, or you just want to kind of give

942
00:53:07,000 --> 00:53:10,279
Speaker 1: a better idea of what's going on when you are

943
00:53:10,480 --> 00:53:14,359
Speaker 1: using a device or a service with Alexa incorporated into

944
00:53:14,400 --> 00:53:17,440
Speaker 1: it and you're wondering, well, what's actually happening? Those like

945
00:53:17,480 --> 00:53:20,879
Speaker 1: I said, those blog posts are really accessible. I've read

946
00:53:21,480 --> 00:53:24,320
Speaker 1: a lot of developer blogs over the last ten years,

947
00:53:24,640 --> 00:53:27,239
Speaker 1: and they are written in a way that is much

948
00:53:27,440 --> 00:53:31,560
Speaker 1: easier to understand, even if you're coming in from a

949
00:53:31,640 --> 00:53:35,200
Speaker 1: purely just just an area of curiosity, much easier to

950
00:53:35,280 --> 00:53:37,480
Speaker 1: understand that some of the other ones I've have encountered.

951
00:53:38,320 --> 00:53:42,719
Speaker 1: That makes me feel good. And I not that you

952
00:53:42,760 --> 00:53:46,000
Speaker 1: know it's I am. I am a budding uh podcaster,

953
00:53:46,120 --> 00:53:49,560
Speaker 1: but just last week I did actually launch a podcast

954
00:53:49,600 --> 00:53:53,400
Speaker 1: along the same lines of just talking about Alexa, you know,

955
00:53:53,440 --> 00:53:56,840
Speaker 1: similar to the blog post learning Um. The first episode

956
00:53:56,880 --> 00:53:59,200
Speaker 1: aired with Charlie Kendall, who runs in our smart home,

957
00:53:59,239 --> 00:54:02,120
Speaker 1: so it's me asking him you know what that team

958
00:54:02,440 --> 00:54:04,640
Speaker 1: set out to accomplish it If you're if you're interested,

959
00:54:04,680 --> 00:54:06,640
Speaker 1: if you're if your listeners want to check it out,

960
00:54:06,680 --> 00:54:10,600
Speaker 1: it's Bitley Alexa dev Chat or just look for Alexa

961
00:54:10,640 --> 00:54:14,680
Speaker 1: dev Chat on iTunes or Stitcher or tune in or

962
00:54:14,719 --> 00:54:18,120
Speaker 1: any of the other catchers. Yeah yeah, I should also

963
00:54:18,160 --> 00:54:22,120
Speaker 1: mention the video series for learning Alexa skill Kit Development

964
00:54:22,840 --> 00:54:26,920
Speaker 1: just rolled out. At least the first two videos have

965
00:54:27,080 --> 00:54:29,520
Speaker 1: rolled out, and um I believe Amazon is planning on

966
00:54:29,640 --> 00:54:33,000
Speaker 1: rolling all of them out in the next few days. Um,

967
00:54:33,040 --> 00:54:34,960
Speaker 1: so those are going to be available online. If you

968
00:54:34,960 --> 00:54:39,000
Speaker 1: search for you know, Big nerd Ranch Amazon training videos,

969
00:54:39,000 --> 00:54:41,480
Speaker 1: you should you should be able to find it that way.

970
00:54:41,480 --> 00:54:45,120
Speaker 1: Fantastic guys. Thank you again and listeners out there. If

971
00:54:45,120 --> 00:54:46,440
Speaker 1: you want to get in touch with me, you have

972
00:54:46,440 --> 00:54:48,919
Speaker 1: a suggestion for a future episode, you gotta follow up question.

973
00:54:48,960 --> 00:54:51,360
Speaker 1: You want me to either run down Dave or Josh

974
00:54:51,360 --> 00:54:53,879
Speaker 1: and ask them, or you know, just want to say hi.

975
00:54:54,400 --> 00:54:57,479
Speaker 1: You can email me addresses tech stuff at how stuff

976
00:54:57,480 --> 00:54:59,760
Speaker 1: works dot com or drop me a line on Twitter

977
00:54:59,880 --> 00:55:02,880
Speaker 1: or Facebook. The handle for both of those is text

978
00:55:02,880 --> 00:55:06,120
Speaker 1: stuff h s W and I'll talk to you again

979
00:55:06,840 --> 00:55:14,960
Speaker 1: really soon. For more on this and thousands of other topics,

980
00:55:15,200 --> 00:55:26,840
Speaker 1: is it how stuff works dot com.