1
00:00:05,920 --> 00:00:09,160
Speaker 1: Hey, welcome to Stuff to Blow Your Mind. This is

2
00:00:09,240 --> 00:00:12,520
Speaker 1: Robert Lamb and Hey it's Saturday. It's time to venture

3
00:00:12,560 --> 00:00:15,480
Speaker 1: into the vault once more for one of our core

4
00:00:15,560 --> 00:00:18,520
Speaker 1: episodes from the past. This is going to be The

5
00:00:18,560 --> 00:00:24,120
Speaker 1: Machine Speaks, Originally published on six twenty nine, twenty twenty three.

6
00:00:24,600 --> 00:00:27,600
Speaker 1: We really hope you enjoy this one. It's a journey

7
00:00:27,680 --> 00:00:32,080
Speaker 1: through everything from ancient tales of artifice and wizardry to

8
00:00:32,159 --> 00:00:37,040
Speaker 1: the early breakthroughs in speech synthesis technology. Let's dive right in.

9
00:00:40,479 --> 00:00:44,640
Speaker 2: Welcome to Stuff to Blow Your Mind, production of iHeartRadio.

10
00:00:50,240 --> 00:00:52,360
Speaker 1: Hey, welcome to Stuff to Blow Your Mind. My name

11
00:00:52,400 --> 00:00:53,000
Speaker 1: is Robert.

12
00:00:52,840 --> 00:00:56,560
Speaker 3: Lamb, and I am Joe McCormick, and today we're going

13
00:00:56,640 --> 00:01:02,160
Speaker 3: to be talking about some early voice synthesis machines. Rob

14
00:01:02,160 --> 00:01:05,440
Speaker 3: I actually got interested in this topic because last week,

15
00:01:05,480 --> 00:01:09,000
Speaker 3: when we were watching the Weird House Cinema movie The

16
00:01:09,040 --> 00:01:12,800
Speaker 3: Black Hole, I was thinking about Roddy McDowell's voice when

17
00:01:12,840 --> 00:01:16,399
Speaker 3: he's doing a voice for the robot character who shares

18
00:01:16,400 --> 00:01:19,280
Speaker 3: a lot of proverbs with the human characters. And I

19
00:01:19,400 --> 00:01:23,400
Speaker 3: kept listening to his line delivery and I couldn't decide

20
00:01:23,440 --> 00:01:26,640
Speaker 3: if he was trying to do quote robot voice or not.

21
00:01:27,040 --> 00:01:28,560
Speaker 3: He seemed to kind of dip in and out of it.

22
00:01:28,600 --> 00:01:30,480
Speaker 3: You know what I mean when I say robot voice,

23
00:01:30,480 --> 00:01:33,480
Speaker 3: where a character's playing a robot and they say things

24
00:01:33,680 --> 00:01:35,040
Speaker 3: like this, Well.

25
00:01:34,880 --> 00:01:37,119
Speaker 1: Of course I know what you are talking about, Joe.

26
00:01:37,760 --> 00:01:41,199
Speaker 3: I got kind of interested in the history of robot voice.

27
00:01:41,280 --> 00:01:43,880
Speaker 3: I was like, where does that come from? And I

28
00:01:43,920 --> 00:01:45,560
Speaker 3: was digging around a little. I'm sure there is a

29
00:01:45,600 --> 00:01:48,200
Speaker 3: good answer on that, but I don't know. My short

30
00:01:48,240 --> 00:01:51,560
Speaker 3: search didn't really turn up anything interesting, but it did

31
00:01:51,760 --> 00:01:55,560
Speaker 3: lead me indirectly to what we're talking about today, which is,

32
00:01:55,680 --> 00:02:00,520
Speaker 3: of course, we have the voice synthesis systems that that

33
00:02:00,560 --> 00:02:03,320
Speaker 3: are largely digital today. Before that, you had a lot

34
00:02:03,400 --> 00:02:09,400
Speaker 3: of electrical and electro mechanical systems for synthesizing human voices.

35
00:02:09,680 --> 00:02:12,359
Speaker 3: But actually there is an even earlier generation, which are

36
00:02:12,400 --> 00:02:17,520
Speaker 3: the purely mechanical voice synthesizers before electricity even came into

37
00:02:17,560 --> 00:02:21,040
Speaker 3: the picture. And that is what really stole my heart,

38
00:02:21,160 --> 00:02:23,920
Speaker 3: especially one particular machine of this type that I'm going

39
00:02:23,960 --> 00:02:25,960
Speaker 3: to talk about in the second half of this episode.

40
00:02:26,000 --> 00:02:29,120
Speaker 1: I think, yeah, this is a fascinating topic, in part

41
00:02:29,200 --> 00:02:31,359
Speaker 1: because look at it. Look at where we are now,

42
00:02:31,520 --> 00:02:34,640
Speaker 1: right it's easy today in our Internet age for just

43
00:02:34,680 --> 00:02:38,640
Speaker 1: the average Internet user to engage with various chatbots and

44
00:02:38,720 --> 00:02:43,760
Speaker 1: generative AI, text to speech and so forth. And so

45
00:02:43,800 --> 00:02:47,560
Speaker 1: we're able to interact with an artifact, a thing that

46
00:02:47,600 --> 00:02:50,480
Speaker 1: reflects human will, that has been designed to do key

47
00:02:50,639 --> 00:02:53,480
Speaker 1: and telling things that have long been the hallmarks of

48
00:02:53,560 --> 00:02:59,800
Speaker 1: human activity, artistic generation, creative writing and conversation, or especially speech.

49
00:03:00,840 --> 00:03:03,240
Speaker 1: And of course it's you know, it's easy nowadays to

50
00:03:03,280 --> 00:03:06,320
Speaker 1: do that, right, to transform into audible or even video

51
00:03:06,440 --> 00:03:11,000
Speaker 1: content what is either written by human or or created

52
00:03:11,040 --> 00:03:15,080
Speaker 1: with some sort of a chat bot machine. And the

53
00:03:15,160 --> 00:03:18,480
Speaker 1: results may be amusing, they may may be disastrous. But

54
00:03:18,639 --> 00:03:20,680
Speaker 1: we're in this age where the idea of the machine

55
00:03:20,720 --> 00:03:24,440
Speaker 1: speaking is not in and of itself groundbreaking, or at

56
00:03:24,520 --> 00:03:28,119
Speaker 1: least if it is groundbreaking, or if it's amazing, it's

57
00:03:28,960 --> 00:03:33,160
Speaker 1: that it's a lower level of amazement compared to previous ages.

58
00:03:33,760 --> 00:03:36,440
Speaker 3: Well, as you say, it's very integrated into modern technology.

59
00:03:36,520 --> 00:03:39,480
Speaker 3: So there's you know, Siri and Alexa, all these like

60
00:03:39,680 --> 00:03:43,520
Speaker 3: home devices that speak, GPS devices for the car, you know,

61
00:03:43,600 --> 00:03:46,480
Speaker 3: that speak to you, but almost all of them are

62
00:03:46,520 --> 00:03:49,320
Speaker 3: still the subject of amusement if you actually pay attention

63
00:03:49,360 --> 00:03:52,320
Speaker 3: to what the voice sounds like, you know, like reading

64
00:03:52,360 --> 00:03:54,800
Speaker 3: emotions into the voice that's telling you what to do

65
00:03:54,840 --> 00:03:57,920
Speaker 3: as you're driving. That always makes me laugh because it

66
00:03:57,920 --> 00:03:59,400
Speaker 3: always seems a little bit annoyed.

67
00:04:00,120 --> 00:04:03,360
Speaker 1: Yeah, yeah, what's series whole deal that sort of thing, right?

68
00:04:05,200 --> 00:04:07,240
Speaker 1: You know. The other interesting angle on all of this

69
00:04:07,600 --> 00:04:12,880
Speaker 1: is that are modern technological advancements here, or even some

70
00:04:12,920 --> 00:04:15,960
Speaker 1: of the historic technological advancements like they are kind of

71
00:04:16,000 --> 00:04:20,040
Speaker 1: the echo of a more ancient longing for this sort

72
00:04:20,040 --> 00:04:25,120
Speaker 1: of thing. It connects to something that's just fascinated us

73
00:04:25,120 --> 00:04:28,359
Speaker 1: for a long time, the idea generally of non human

74
00:04:28,520 --> 00:04:31,560
Speaker 1: entities engaging in speech. And you could you could go

75
00:04:31,600 --> 00:04:35,280
Speaker 1: absolutely wild chasing down the various divisions of this, right,

76
00:04:35,360 --> 00:04:40,160
Speaker 1: the various myths, legends, and traditions concerning the speech of animals, plants,

77
00:04:40,560 --> 00:04:47,080
Speaker 1: inorganic materials, supernatural entities, you know, voices seemingly internal but

78
00:04:47,160 --> 00:04:49,920
Speaker 1: also external to our individual experience.

79
00:04:50,400 --> 00:04:52,599
Speaker 3: Though I would say there is an interesting thing about

80
00:04:52,800 --> 00:04:57,120
Speaker 3: machines or human or automata or human artifacts in general

81
00:04:57,440 --> 00:05:01,920
Speaker 3: when compared to imagining an animal speaker or any other

82
00:05:02,279 --> 00:05:05,080
Speaker 3: usually not speaking things starting to speak, which is, if

83
00:05:05,120 --> 00:05:07,799
Speaker 3: you're talking about a machine that does it. That means

84
00:05:07,960 --> 00:05:10,839
Speaker 3: somebody has to make that machine and somebody has to

85
00:05:10,920 --> 00:05:13,560
Speaker 3: work that machine, and it kind of reminds me of

86
00:05:13,680 --> 00:05:16,960
Speaker 3: the idea of grammar in language. You know, the interesting

87
00:05:16,960 --> 00:05:20,080
Speaker 3: thing about grammar is that when we use language, we

88
00:05:20,120 --> 00:05:23,279
Speaker 3: all use grammar, so we have an intuitive grasp of

89
00:05:23,320 --> 00:05:26,760
Speaker 3: the rules of grammar, but without serious study, people can't

90
00:05:26,760 --> 00:05:29,000
Speaker 3: actually tell you what those rules are. And so, like,

91
00:05:29,040 --> 00:05:31,000
Speaker 3: you know, that had to be in a sense a

92
00:05:31,279 --> 00:05:34,919
Speaker 3: science to back engineer the rules of grammar that we

93
00:05:35,000 --> 00:05:38,680
Speaker 3: use intuitively to like make them systematic and you know,

94
00:05:38,800 --> 00:05:42,000
Speaker 3: actually discover what those rules are. The same thing could

95
00:05:42,040 --> 00:05:45,880
Speaker 3: be said about the phonetic rules that produce the intelligible speech.

96
00:05:46,279 --> 00:05:48,839
Speaker 3: We can all do it if we can speak, but

97
00:05:49,120 --> 00:05:54,160
Speaker 3: we don't necessarily understand what the individual physical properties of

98
00:05:54,200 --> 00:05:57,200
Speaker 3: a word are, and so we wouldn't necessarily know how

99
00:05:57,240 --> 00:05:59,800
Speaker 3: to make that same word come out of a machine.

100
00:06:00,600 --> 00:06:02,520
Speaker 1: Yeah, there are all these things that you have to

101
00:06:02,560 --> 00:06:06,200
Speaker 1: deconstruct before you can attempt to reproduce it artificially. And

102
00:06:06,279 --> 00:06:09,440
Speaker 1: we see that time and time again with in robotics,

103
00:06:09,440 --> 00:06:13,000
Speaker 1: for example. You know, things that we take for granted

104
00:06:13,320 --> 00:06:17,000
Speaker 1: concerning human movement, just about anything else you could imagine

105
00:06:17,080 --> 00:06:19,600
Speaker 1: it becomes so much more difficult to try and reproduce

106
00:06:19,640 --> 00:06:22,479
Speaker 1: that you've got to understand what it actually is on

107
00:06:22,520 --> 00:06:23,640
Speaker 1: an entirely new level.

108
00:06:23,680 --> 00:06:27,480
Speaker 3: First, Now, I am to understand that before anybody actually

109
00:06:27,520 --> 00:06:30,760
Speaker 3: made a machine that could approximate or synthesize a human

110
00:06:30,839 --> 00:06:34,479
Speaker 3: voice and produce intelligible speech, people were thinking about this

111
00:06:34,600 --> 00:06:35,480
Speaker 3: as a concept.

112
00:06:36,360 --> 00:06:38,680
Speaker 1: Yeah yeah, and this is not surprising. You know, this

113
00:06:38,880 --> 00:06:43,559
Speaker 1: is kind of the meat of science fiction. Right before

114
00:06:43,560 --> 00:06:45,839
Speaker 1: we can do it, we dream of it one way

115
00:06:45,960 --> 00:06:49,679
Speaker 1: or another, no matter what our exact grasp of science

116
00:06:49,720 --> 00:06:53,080
Speaker 1: happens to be. It always reminds me of that line

117
00:06:53,120 --> 00:06:57,880
Speaker 1: in William Gibson's Neuromancer where the character has made a deal,

118
00:06:57,960 --> 00:07:01,320
Speaker 1: a pact with a powerful AI, and it's pointed out like,

119
00:07:01,360 --> 00:07:04,000
Speaker 1: this is the sort of thing that in you know,

120
00:07:04,080 --> 00:07:07,200
Speaker 1: centuries ago, people only dreamed of making a deal with

121
00:07:07,320 --> 00:07:10,120
Speaker 1: a devil, and now we've made it possible through our

122
00:07:10,280 --> 00:07:11,320
Speaker 1: ingenuity and invention.

123
00:07:11,760 --> 00:07:12,880
Speaker 3: Congratulations.

124
00:07:13,240 --> 00:07:18,360
Speaker 1: Yeah, so so yeah. Narrowing down here into generally the

125
00:07:18,360 --> 00:07:27,200
Speaker 1: realm of alleged human creations that through at least partial technology,

126
00:07:27,280 --> 00:07:30,800
Speaker 1: but also sometimes wizardry and alchemy and other things that

127
00:07:30,840 --> 00:07:33,360
Speaker 1: are kind of like you know, bunched in there together

128
00:07:33,520 --> 00:07:37,760
Speaker 1: with with actual technology to create some sort of a

129
00:07:37,840 --> 00:07:40,600
Speaker 1: device capable of speech. And then there are some also

130
00:07:40,720 --> 00:07:42,880
Speaker 1: some related things that are tied in there as well.

131
00:07:43,080 --> 00:07:44,960
Speaker 1: And a lot of it comes down to the idea

132
00:07:45,320 --> 00:07:49,840
Speaker 1: of a head, an artificial head that speaks.

133
00:07:50,480 --> 00:07:53,800
Speaker 3: I found something so loaded and revealing about that. As

134
00:07:53,840 --> 00:07:56,800
Speaker 3: a fact, the history of these machines, so many of

135
00:07:56,840 --> 00:08:00,680
Speaker 3: them had fa whether real or imagined, These machines, so

136
00:08:00,800 --> 00:08:04,800
Speaker 3: many of them early on had heads or faces, so

137
00:08:04,840 --> 00:08:07,000
Speaker 3: like it wouldn't just be a speaker like you would

138
00:08:07,000 --> 00:08:10,000
Speaker 3: have today, that's you know, it's just a mechanical device

139
00:08:10,040 --> 00:08:13,040
Speaker 3: for making the sound. It's like that the presence of

140
00:08:13,080 --> 00:08:16,320
Speaker 3: a head or a face was considered important or at

141
00:08:16,360 --> 00:08:17,240
Speaker 3: least desirable.

142
00:08:18,000 --> 00:08:19,880
Speaker 1: Yeah, and I wondered to what extent part of it

143
00:08:19,920 --> 00:08:24,000
Speaker 1: is just an echo of these earlier ideas. So going

144
00:08:24,040 --> 00:08:27,680
Speaker 1: to run through a few of these here. One of

145
00:08:27,720 --> 00:08:31,400
Speaker 1: the most famous, mainly from a literary tradition, as we'll

146
00:08:31,520 --> 00:08:35,000
Speaker 1: discuss here, is the idea of the brazen head. And

147
00:08:35,440 --> 00:08:37,480
Speaker 1: ultimately I guess there's more than one brazen head. We

148
00:08:37,520 --> 00:08:42,720
Speaker 1: can say brazen heads artificial heads that could speak. There's

149
00:08:42,800 --> 00:08:46,400
Speaker 1: a basically a lot of these stories concerned thirteenth century

150
00:08:46,440 --> 00:08:50,600
Speaker 1: English philosopher and Franciscan friar Roger Bacon, who's come up

151
00:08:50,600 --> 00:08:54,360
Speaker 1: on the show before, though this particular version of the

152
00:08:54,360 --> 00:08:57,440
Speaker 1: story doesn't seem to emerge until the sixteenth century, and

153
00:08:57,520 --> 00:09:00,680
Speaker 1: it does so within the works of contemporary drama.

154
00:09:01,160 --> 00:09:03,400
Speaker 3: I think we talked about Roger Bacon at length in

155
00:09:03,440 --> 00:09:06,360
Speaker 3: an episode we did about the invention of fireworks, which

156
00:09:07,000 --> 00:09:09,360
Speaker 3: may come back and feature again in the feed soon.

157
00:09:10,040 --> 00:09:12,000
Speaker 1: Yeah, yeah, I believe you're right. Yeah. I think Bacon

158
00:09:12,000 --> 00:09:15,199
Speaker 1: did come up in that he had a reputation as

159
00:09:15,280 --> 00:09:18,960
Speaker 1: not only a very learned man in both natural philosophy

160
00:09:19,000 --> 00:09:22,319
Speaker 1: and theology, and I should drive home definitely existed. I

161
00:09:22,360 --> 00:09:24,560
Speaker 1: don't think there's any doubt that there was a Roger Bacon.

162
00:09:25,080 --> 00:09:27,280
Speaker 1: But then there are all these other stories that he

163
00:09:27,400 --> 00:09:31,880
Speaker 1: was also potentially a wizard who was capable of producing

164
00:09:32,000 --> 00:09:37,880
Speaker 1: fabulous automata, either through amazing feats of clockwork ingenuity that

165
00:09:37,960 --> 00:09:42,280
Speaker 1: I think many would say was ultimately, you know, impossible

166
00:09:42,360 --> 00:09:46,600
Speaker 1: during his time period, or failing that, he was into

167
00:09:46,679 --> 00:09:49,320
Speaker 1: alchemy and of course dark dank necromancy.

168
00:09:49,960 --> 00:09:52,000
Speaker 3: I think the way I conceive of Roger Bacon is

169
00:09:52,040 --> 00:09:55,360
Speaker 3: that he of course was a real figure. He was

170
00:09:56,120 --> 00:10:00,520
Speaker 3: of great intellectual note and significance, but much about his

171
00:10:00,600 --> 00:10:04,280
Speaker 3: sort of general reputation is kind of legendary, if that

172
00:10:04,320 --> 00:10:05,840
Speaker 3: makes sense. I mean, there are many things we know

173
00:10:05,920 --> 00:10:09,559
Speaker 3: about him that are true, but there's also just sort

174
00:10:09,600 --> 00:10:12,160
Speaker 3: of an aura or a vibe about him that is

175
00:10:12,200 --> 00:10:13,560
Speaker 3: not really based in reality.

176
00:10:14,280 --> 00:10:17,360
Speaker 1: Yeah, I mean, he becomes a character in literature, especially

177
00:10:17,360 --> 00:10:19,680
Speaker 1: in these accounts, so you can you can sort of

178
00:10:19,920 --> 00:10:23,320
Speaker 1: look at the different phases like historic individual ideas and

179
00:10:23,360 --> 00:10:27,880
Speaker 1: you know, misunderstandings have said real life individual and then

180
00:10:27,920 --> 00:10:31,320
Speaker 1: eventually that echoes into the fictional version of the person.

181
00:10:31,360 --> 00:10:33,400
Speaker 3: Which that's the more like wizard version.

182
00:10:34,120 --> 00:10:36,800
Speaker 1: Yeah, yeah, And so there are a few different examples

183
00:10:36,840 --> 00:10:39,160
Speaker 1: of this. This was like a popular motif for a while.

184
00:10:39,559 --> 00:10:43,480
Speaker 1: There's a sixteenth century prose romance titled the Famous History

185
00:10:43,480 --> 00:10:45,920
Speaker 1: of Friar Bacon, and it tells of Bacon trying to

186
00:10:46,400 --> 00:10:48,920
Speaker 1: give a replica of a human head speech and having

187
00:10:48,960 --> 00:10:52,800
Speaker 1: to call in the devil for help. Cool. Other versions

188
00:10:52,800 --> 00:10:54,959
Speaker 1: of this tale describe it as an artificial head given

189
00:10:55,040 --> 00:10:58,920
Speaker 1: life by demons, which was capable of spontaneous speech and

190
00:10:58,960 --> 00:11:02,840
Speaker 1: of course telling the future I mean, what else would

191
00:11:02,880 --> 00:11:08,160
Speaker 1: you tell right right? Robert Greene's sixteen thirty play Friar

192
00:11:08,240 --> 00:11:12,880
Speaker 1: Bacon and Friar Bungay mentions this several times, citing quote

193
00:11:13,000 --> 00:11:18,199
Speaker 1: Bacon's necromatic skill and heads of Brass that quote can

194
00:11:18,280 --> 00:11:21,160
Speaker 1: utter any voice. The idea that's exploring both of these

195
00:11:21,200 --> 00:11:23,679
Speaker 1: works is that Bacon wished to build a wall of

196
00:11:23,720 --> 00:11:27,000
Speaker 1: brass around Britain with the help of the Brazen Head.

197
00:11:27,720 --> 00:11:29,640
Speaker 1: He fails and the head explodes.

198
00:11:32,080 --> 00:11:33,920
Speaker 3: Why have I never heard this side? It as like

199
00:11:33,960 --> 00:11:35,640
Speaker 3: an early science fiction tale.

200
00:11:36,120 --> 00:11:38,120
Speaker 1: I don't know. I'm probably not doing due diligence on

201
00:11:38,280 --> 00:11:41,040
Speaker 1: exactly what happens and everything, at least like saying, well,

202
00:11:41,080 --> 00:11:43,719
Speaker 1: in Star Wars, the bad guys make one planet to

203
00:11:43,720 --> 00:11:45,920
Speaker 1: blow up another planet, and then the planet they may

204
00:11:46,000 --> 00:11:48,760
Speaker 1: blows up. You know, That's that's really skipping over a

205
00:11:48,760 --> 00:11:51,040
Speaker 1: lot of the nuance. And so I think there's there's

206
00:11:51,080 --> 00:11:53,280
Speaker 1: inevitably more nuance here, but I just I didn't get

207
00:11:53,280 --> 00:11:57,280
Speaker 1: into it. Okay, So this idea of the satanic brasshead

208
00:11:57,280 --> 00:12:00,480
Speaker 1: of Roger Bacon persists despite the fact that there's no

209
00:12:00,520 --> 00:12:04,520
Speaker 1: indication that anything like this even created purely through technology

210
00:12:04,559 --> 00:12:08,160
Speaker 1: and not Satanic wizardry was part of Bacon's world. He

211
00:12:08,280 --> 00:12:12,720
Speaker 1: was interested in optics and certainly various instruments scientific instruments

212
00:12:12,720 --> 00:12:14,680
Speaker 1: of brass of the day. But there's no indication that

213
00:12:14,720 --> 00:12:17,080
Speaker 1: he ever built an artificial head and tried to get

214
00:12:17,120 --> 00:12:17,760
Speaker 1: it to speak.

215
00:12:18,040 --> 00:12:20,320
Speaker 3: Okay, so this is part of the wizard aura, not

216
00:12:20,400 --> 00:12:22,360
Speaker 3: part of his biography.

217
00:12:22,280 --> 00:12:25,000
Speaker 1: Right though, you know, we have to drive home to

218
00:12:25,240 --> 00:12:28,600
Speaker 1: is it possible that Roger Bacon, as a hobby did

219
00:12:28,640 --> 00:12:31,040
Speaker 1: what he could to create you know, I mean it's possible,

220
00:12:31,080 --> 00:12:33,280
Speaker 1: it's not. You know, I don't think he would have

221
00:12:33,400 --> 00:12:36,120
Speaker 1: gotten to speak. But there are various sort of ways

222
00:12:36,160 --> 00:12:38,880
Speaker 1: you could interpret this as having some basis in reality

223
00:12:39,320 --> 00:12:43,480
Speaker 1: that doesn't involve magic or superscience of the day. Okay, Now,

224
00:12:43,640 --> 00:12:46,000
Speaker 1: I went to my bookshelf and I pulled off my

225
00:12:46,120 --> 00:12:49,680
Speaker 1: dusty copy of Brewer's Dictionary Phrase and Fable. It provides

226
00:12:49,679 --> 00:12:52,760
Speaker 1: a little more insight on the legend. Quote. It was

227
00:12:52,800 --> 00:12:55,960
Speaker 1: said if Bacon heard it speak, he would succeed in

228
00:12:55,960 --> 00:12:59,920
Speaker 1: his projects, if not, he would fail. His familiar mile

229
00:13:00,280 --> 00:13:03,000
Speaker 1: was set to watch, and while Bacon slept, the head

230
00:13:03,040 --> 00:13:07,400
Speaker 1: spoke thrice. Time is half an hour later, it said,

231
00:13:07,760 --> 00:13:11,840
Speaker 1: time was in another half hour, it said times past

232
00:13:12,280 --> 00:13:16,640
Speaker 1: fell down and was broken to atoms to atoms to atoms. Yes,

233
00:13:17,440 --> 00:13:21,160
Speaker 1: surely Adams means something different here, Adams rights been discovered

234
00:13:21,200 --> 00:13:23,600
Speaker 1: at the time. I think it just means like small

235
00:13:23,640 --> 00:13:24,520
Speaker 1: parts or something.

236
00:13:24,640 --> 00:13:28,400
Speaker 3: Yeah, yes, yeah, that would be hilarious if it was

237
00:13:28,480 --> 00:13:29,960
Speaker 3: literally broken to atoms.

238
00:13:30,600 --> 00:13:32,280
Speaker 1: Yeah. So I don't know if it if it works though,

239
00:13:32,320 --> 00:13:34,720
Speaker 1: it sounds like it's kind of an alarm clock that explodes.

240
00:13:35,080 --> 00:13:38,560
Speaker 3: Well, but I don't understand the difference between time was

241
00:13:38,800 --> 00:13:41,360
Speaker 3: and times pasted. They're both past tens.

242
00:13:41,920 --> 00:13:44,400
Speaker 1: Hm m. That's a good point. Time is, time was

243
00:13:45,080 --> 00:13:47,840
Speaker 1: times past. It seems like you would want the president

244
00:13:47,920 --> 00:13:52,439
Speaker 1: there somewhere, but yeah, that's that's that's what it allegedly said.

245
00:13:52,840 --> 00:13:57,679
Speaker 1: And you'll you'll find woodcuts that that have this this

246
00:13:57,840 --> 00:13:59,240
Speaker 1: motif on them as well.

247
00:13:59,600 --> 00:14:01,160
Speaker 3: Like it would it made more sense if it said

248
00:14:01,679 --> 00:14:05,000
Speaker 3: the three things where time will be, time is, time was,

249
00:14:05,920 --> 00:14:08,440
Speaker 3: but this seems more like time is, time was, time

250
00:14:08,679 --> 00:14:09,160
Speaker 3: was was.

251
00:14:11,240 --> 00:14:14,360
Speaker 1: Now. Brewers notes that reference to the references to the

252
00:14:14,360 --> 00:14:17,840
Speaker 1: Brazen Head are just common in literature, appearing frequently in

253
00:14:17,880 --> 00:14:23,080
Speaker 1: early romances but with Eastern origins, though it doesn't get

254
00:14:23,080 --> 00:14:26,200
Speaker 1: into that a lot elsewhere in the volume. It's also

255
00:14:26,320 --> 00:14:32,040
Speaker 1: noted that artificial heads that speak occur elsewhere as well.

256
00:14:32,080 --> 00:14:33,880
Speaker 1: And some of these are brazen heads, and some of

257
00:14:33,960 --> 00:14:35,680
Speaker 1: these are other things, but they're kind of I think

258
00:14:35,680 --> 00:14:38,440
Speaker 1: it's important to run through briefly some of these examples

259
00:14:38,480 --> 00:14:41,080
Speaker 1: because they kind of paint a picture of not only

260
00:14:41,120 --> 00:14:44,600
Speaker 1: some of these other ideas of artificial heads speaking and

261
00:14:44,640 --> 00:14:52,240
Speaker 1: telling the future, but related non technological non artifacts that

262
00:14:52,640 --> 00:14:57,960
Speaker 1: kind of help inform what we think technology can do. Okay, okay,

263
00:14:58,000 --> 00:14:59,920
Speaker 1: So one of them is a brazen head in the

264
00:15:00,000 --> 00:15:02,960
Speaker 1: possession of Pope Sylvester the Second in the tenth century,

265
00:15:03,240 --> 00:15:07,680
Speaker 1: which he also constructed, and misinterpretations of its utterances could

266
00:15:07,760 --> 00:15:08,720
Speaker 1: prove disastrous.

267
00:15:09,200 --> 00:15:13,280
Speaker 3: Oh, is this also believed to be Satanic in some way?

268
00:15:14,600 --> 00:15:19,920
Speaker 1: I didn't go too deep on Satanic implications, but possibly, I.

269
00:15:19,880 --> 00:15:22,240
Speaker 3: Guess it would depend on if this legend is associated

270
00:15:22,280 --> 00:15:25,920
Speaker 3: with pro Pope Sylvester or anti Pope Sylvester sources.

271
00:15:26,040 --> 00:15:28,680
Speaker 1: Right, right, But you can definitely see that they're in

272
00:15:28,720 --> 00:15:31,040
Speaker 1: the head itself, regardless of what's supposed to be powering it.

273
00:15:31,120 --> 00:15:34,160
Speaker 1: Like this, it ties into two oracular traditions. You know,

274
00:15:34,240 --> 00:15:37,320
Speaker 1: the idea that here is this thing that can give

275
00:15:37,400 --> 00:15:41,120
Speaker 1: you cryptic wisdom if you have the wisdom to decipher

276
00:15:41,160 --> 00:15:43,760
Speaker 1: what it's telling you. Another example that's brought up in

277
00:15:43,800 --> 00:15:48,560
Speaker 1: Brewers is or the Colossi of Memnon, which we did

278
00:15:48,640 --> 00:15:50,240
Speaker 1: at least a whole I don't know, I can't remember

279
00:15:50,240 --> 00:15:52,560
Speaker 1: as one episode or multiple episodes, but we discussed this

280
00:15:53,080 --> 00:15:56,280
Speaker 1: on stuff to blow your mind. This is a fascinating

281
00:15:56,280 --> 00:15:57,400
Speaker 1: topic in and of itself.

282
00:15:57,640 --> 00:15:59,920
Speaker 3: This was basically, I think a statue or a pair

283
00:15:59,920 --> 00:16:03,320
Speaker 3: of statues, part of sort of a ruins complex that

284
00:16:03,520 --> 00:16:08,200
Speaker 3: was famous during in Roman Egypt as basically because it

285
00:16:08,240 --> 00:16:11,280
Speaker 3: would make sounds, and there were different theories about how

286
00:16:11,320 --> 00:16:12,600
Speaker 3: it made sounds and why.

287
00:16:12,920 --> 00:16:15,920
Speaker 1: Yeah, yeah, it seems like I think some said it

288
00:16:15,960 --> 00:16:18,320
Speaker 1: was capable of speech, but generally it's described as singing

289
00:16:18,440 --> 00:16:21,600
Speaker 1: or some sort of a note. And as we discussed,

290
00:16:21,840 --> 00:16:24,840
Speaker 1: while there are some I think unlikely theories regarding the

291
00:16:24,920 --> 00:16:28,640
Speaker 1: use of some sort of intentional sound generating device or devices,

292
00:16:28,920 --> 00:16:31,400
Speaker 1: it seems like a more likely explanation would have to

293
00:16:31,400 --> 00:16:34,880
Speaker 1: do with peculiarities of the stone as it heated in

294
00:16:34,960 --> 00:16:37,880
Speaker 1: the sun and then cooled at night. Anyway, go back

295
00:16:37,920 --> 00:16:39,200
Speaker 1: and listen to that episode if you want to know

296
00:16:39,200 --> 00:16:41,560
Speaker 1: about them. They have a pretty fascinating history.

297
00:16:42,040 --> 00:16:43,640
Speaker 3: We'll remember better in the original.

298
00:16:44,440 --> 00:16:49,480
Speaker 1: Yes, there's the head of Orpheus at Lesbos, predicting the

299
00:16:49,520 --> 00:16:53,080
Speaker 1: doom and death of Cyrus the Great. However, I believe

300
00:16:53,120 --> 00:16:55,400
Speaker 1: this is generally thought to be the actual head of

301
00:16:55,440 --> 00:16:58,280
Speaker 1: the hero Orpheus, after he was torn apart by the

302
00:16:58,280 --> 00:17:02,480
Speaker 1: main ads of Dionysus during a bacchanalia for the sin

303
00:17:02,560 --> 00:17:06,000
Speaker 1: of worshiping Apollo or having worshiped Apollo. I'm not sure

304
00:17:06,040 --> 00:17:09,400
Speaker 1: what the exact charge was, but still a prophetic, disembodied

305
00:17:09,440 --> 00:17:13,199
Speaker 1: head that still continues to speak. Brewers also mentions the

306
00:17:13,240 --> 00:17:17,439
Speaker 1: head of Minos brought by Odin to Scandinavia, which I

307
00:17:17,480 --> 00:17:20,040
Speaker 1: didn't know what to make of this, because Minos is

308
00:17:20,119 --> 00:17:26,840
Speaker 1: of course the mythical king of Crete that we've discussed

309
00:17:26,840 --> 00:17:29,320
Speaker 1: on the show before as well. I think the actual

310
00:17:29,480 --> 00:17:33,040
Speaker 1: figure in reference here might be Nimir, the god of

311
00:17:33,080 --> 00:17:39,080
Speaker 1: wisdom that is beheaded in the Aservaniir War. Odin claims

312
00:17:39,080 --> 00:17:42,240
Speaker 1: this head and it continues to speak secret wisdom. Again,

313
00:17:42,320 --> 00:17:44,560
Speaker 1: this is another one that's not a mechanical head. It's

314
00:17:44,560 --> 00:17:48,160
Speaker 1: the head of an actual defeated divine being that continues

315
00:17:48,200 --> 00:17:51,840
Speaker 1: to live on and to speak. There are tales of

316
00:17:51,920 --> 00:17:55,760
Speaker 1: Albertus Magnus having an earthen head, which during the thirteenth

317
00:17:55,840 --> 00:17:59,760
Speaker 1: century was said to speak and move until Thomas Aquinas

318
00:17:59,760 --> 00:18:02,720
Speaker 1: breaks sit by accident, and Magnus says, there goes the

319
00:18:02,760 --> 00:18:07,560
Speaker 1: labor of thirty years, because now it's broken. So I

320
00:18:07,560 --> 00:18:09,880
Speaker 1: don't know what to make of that one either completely.

321
00:18:09,920 --> 00:18:13,359
Speaker 1: But again we see this motif of a fabulous artificial

322
00:18:13,359 --> 00:18:16,919
Speaker 1: head that speaks, that manages to break one way or another,

323
00:18:17,000 --> 00:18:21,280
Speaker 1: either something fails, somebody knocks it over, or you know,

324
00:18:21,359 --> 00:18:24,840
Speaker 1: it explodes after you hit this nooze alarm twice. Then

325
00:18:24,880 --> 00:18:29,000
Speaker 1: there's Alexander's statue of Ascalapius, the Greek god of medicine,

326
00:18:29,240 --> 00:18:32,840
Speaker 1: that was said to speak, but Lucian wrote that the

327
00:18:32,880 --> 00:18:37,119
Speaker 1: sounds came via a concealed man who spoke through tubes.

328
00:18:37,720 --> 00:18:42,560
Speaker 1: So here's an example of some sort of of a creation.

329
00:18:43,800 --> 00:18:45,240
Speaker 1: I guess it depends on you look at either a

330
00:18:45,240 --> 00:18:50,920
Speaker 1: statue that isn't intended to speak, or through supernatural machinations speaks,

331
00:18:51,040 --> 00:18:54,240
Speaker 1: But according to Lucian, it's in neither of those. It's

332
00:18:54,240 --> 00:18:56,720
Speaker 1: just tubes and some guy like hiding in the bushes

333
00:18:56,760 --> 00:18:59,840
Speaker 1: speaking through the tubes, which is still clever and still technological,

334
00:19:00,760 --> 00:19:01,680
Speaker 1: but is trickery.

335
00:19:02,359 --> 00:19:05,879
Speaker 3: Nonetheless, I think the Lucian you're alluding to there is

336
00:19:05,960 --> 00:19:07,080
Speaker 3: Lucian of Samosada.

337
00:19:07,280 --> 00:19:09,760
Speaker 1: Is that right? I believe so, yes, yeah, this.

338
00:19:09,800 --> 00:19:12,760
Speaker 3: Is like this was an ancient satirist from Syria who

339
00:19:12,960 --> 00:19:17,560
Speaker 3: is quite hilarious and was kind of a skeptic debunker

340
00:19:17,680 --> 00:19:20,399
Speaker 3: of the of like the second century CE, which is

341
00:19:20,840 --> 00:19:24,119
Speaker 3: sort of strange, but he was in that mold and

342
00:19:24,160 --> 00:19:28,560
Speaker 3: he made like vicious mockery of people of all sorts

343
00:19:28,600 --> 00:19:32,240
Speaker 3: and different philosophies and stuff, and also wrote a satire

344
00:19:32,359 --> 00:19:34,840
Speaker 3: that some people have considered one of the earliest forms

345
00:19:34,840 --> 00:19:35,560
Speaker 3: of science fiction.

346
00:19:37,080 --> 00:19:39,480
Speaker 1: Now this also reminds me this is not I mean,

347
00:19:39,640 --> 00:19:43,000
Speaker 1: I guess it memory series. Maybe it did speak, But

348
00:19:43,960 --> 00:19:46,800
Speaker 1: there was of course the man faced Serpent God Glicon

349
00:19:47,320 --> 00:19:50,440
Speaker 1: of the second century that is often held up as

350
00:19:50,480 --> 00:19:54,120
Speaker 1: being a hoax, like it was actually a puppet according

351
00:19:54,119 --> 00:19:59,360
Speaker 1: to commentators. But I've always wondered what to make of that,

352
00:19:59,440 --> 00:20:03,560
Speaker 1: because it kind of if someone is performing puppetry and

353
00:20:03,560 --> 00:20:07,000
Speaker 1: people are having an emotional or even religious reaction to it,

354
00:20:07,000 --> 00:20:09,280
Speaker 1: it kind of depends how it's presented. Right, are you

355
00:20:09,320 --> 00:20:13,240
Speaker 1: presenting Glican the man face serpent as like, this is it?

356
00:20:13,400 --> 00:20:16,520
Speaker 1: This is an actual man face serpent God, Come take

357
00:20:16,520 --> 00:20:19,760
Speaker 1: a look that it's life. Is proof that he is real?

358
00:20:20,480 --> 00:20:23,680
Speaker 1: Or is it something else? Is it more like performance

359
00:20:23,840 --> 00:20:26,359
Speaker 1: or is it more like reinterpretation? You know, because you

360
00:20:26,400 --> 00:20:30,920
Speaker 1: have plenty of examples where people will carry out performances

361
00:20:30,920 --> 00:20:34,040
Speaker 1: in which people dress as divine and semi divine figures.

362
00:20:34,600 --> 00:20:37,240
Speaker 1: It's not supposed to be like, look at the proof here,

363
00:20:37,440 --> 00:20:40,600
Speaker 1: here is this hero on the stage. This means God

364
00:20:40,680 --> 00:20:41,880
Speaker 1: is real. Funny enough.

365
00:20:41,880 --> 00:20:44,440
Speaker 3: I think Glicon was also written about by Lucian of

366
00:20:44,480 --> 00:20:49,080
Speaker 3: some Asada. But I guess the crucial question is like

367
00:20:49,720 --> 00:20:51,840
Speaker 3: is there an attempt at trickery or not? Like do

368
00:20:52,720 --> 00:20:55,960
Speaker 3: you want the audience to believe there is not somebody

369
00:20:56,080 --> 00:20:56,960
Speaker 3: behind the mask?

370
00:20:57,480 --> 00:21:00,640
Speaker 1: Right? And you know that's interesting because that still kind

371
00:21:00,640 --> 00:21:03,080
Speaker 1: of applies to a lot of what's going on in

372
00:21:03,119 --> 00:21:06,000
Speaker 1: the world today with things like like chat box and

373
00:21:06,040 --> 00:21:09,639
Speaker 1: so forth. And you know, this idea that if we

374
00:21:10,400 --> 00:21:13,480
Speaker 1: know what is coming out of the box, what is

375
00:21:13,560 --> 00:21:17,200
Speaker 1: coming out of the artificial head? And you know, we

376
00:21:17,640 --> 00:21:20,520
Speaker 1: how are we interpreting it? And are we thinking there

377
00:21:20,560 --> 00:21:23,919
Speaker 1: is something there that is not. So it's like, on

378
00:21:23,960 --> 00:21:26,359
Speaker 1: what level is there trickery, and then there is like

379
00:21:26,560 --> 00:21:30,920
Speaker 1: interpretation of the trickery and so forth. But at any rate,

380
00:21:31,080 --> 00:21:32,920
Speaker 1: I think, you know, some of these examples they proved

381
00:21:32,960 --> 00:21:35,520
Speaker 1: that well before people could make any kind of a

382
00:21:35,560 --> 00:21:38,359
Speaker 1: mechanical thing, be it a head or not ahead that

383
00:21:38,440 --> 00:21:41,159
Speaker 1: could speak, we were still capable of dreaming about it.

384
00:21:41,560 --> 00:21:44,080
Speaker 1: And I think there's ample evidence that long before anyone

385
00:21:44,119 --> 00:21:46,919
Speaker 1: attempted to make a head that could talk through mechanical means,

386
00:21:47,400 --> 00:21:52,159
Speaker 1: individuals sought and sometimes found a voice emerging from disembodied heads,

387
00:21:52,440 --> 00:21:55,240
Speaker 1: either real ones the you know, the the remains of

388
00:21:55,280 --> 00:22:00,000
Speaker 1: human beings or other animals, or or likenesses of human heads,

389
00:22:00,880 --> 00:22:04,560
Speaker 1: either attached or detached from statues, and so forth. And

390
00:22:04,800 --> 00:22:07,600
Speaker 1: I think there's room between trickery and belief for the

391
00:22:07,600 --> 00:22:18,199
Speaker 1: suspension of belief and ritual as well to take into account.

392
00:22:19,840 --> 00:22:22,560
Speaker 3: But of course, later on people would end up building real,

393
00:22:22,760 --> 00:22:27,000
Speaker 3: operable machines that were at least attempting to produce speech

394
00:22:27,119 --> 00:22:28,920
Speaker 3: that could be understood by humans.

395
00:22:29,400 --> 00:22:31,560
Speaker 1: That's right, And this is where we get more into

396
00:22:31,640 --> 00:22:36,040
Speaker 1: the deconstruction of what human speech is, which in and

397
00:22:36,040 --> 00:22:41,400
Speaker 1: of itself is a whole subject, but there are key

398
00:22:41,480 --> 00:22:44,440
Speaker 1: moments where we see some major advancements being made here.

399
00:22:44,960 --> 00:22:48,080
Speaker 1: So another major entry to discuss in all of this

400
00:22:48,160 --> 00:22:50,919
Speaker 1: is the work of German born Russian doctor, physicist and

401
00:22:50,960 --> 00:22:55,760
Speaker 1: engineer Christian Gottlieb Kratzenstein, who lives seventeen twenty three through

402
00:22:55,840 --> 00:23:00,280
Speaker 1: seventeen ninety five. So he was a man of various

403
00:22:59,800 --> 00:23:03,639
Speaker 1: including the use of electricity and medicine, and at the

404
00:23:03,640 --> 00:23:06,640
Speaker 1: Saint Petersburg Science Academy at one point offered a prize

405
00:23:06,640 --> 00:23:12,000
Speaker 1: for advancements made in researching the mechanisms behind the vowels AEI, O,

406
00:23:12,080 --> 00:23:15,520
Speaker 1: and you in human speech. So in seventeen seventy nine

407
00:23:15,560 --> 00:23:21,320
Speaker 1: he presented his vowel organ to the university. The vowel

408
00:23:21,400 --> 00:23:25,480
Speaker 1: organ consisted of a series of resonators that produced vowel

409
00:23:25,560 --> 00:23:30,440
Speaker 1: like sounds on a constant pitch when excited by a read.

410
00:23:31,240 --> 00:23:36,119
Speaker 1: I found some illustrations of these basic resonators via the

411
00:23:36,240 --> 00:23:44,000
Speaker 1: UCL Psychology and Language Sciences Department. Here I also found

412
00:23:44,880 --> 00:23:49,280
Speaker 1: a website linked at this website where you can find

413
00:23:49,280 --> 00:23:52,760
Speaker 1: instructions for how to make your own resonators out of

414
00:23:52,800 --> 00:23:56,879
Speaker 1: plumbing supplies, which I found rather insightful. I did not

415
00:23:56,920 --> 00:24:01,560
Speaker 1: attempt it, but if you're into plumbing supplies and vowel sounds,

416
00:24:01,800 --> 00:24:04,000
Speaker 1: it seems like a natural craft choice.

417
00:24:04,280 --> 00:24:07,399
Speaker 3: But the key insight being here that by changing the

418
00:24:07,520 --> 00:24:10,919
Speaker 3: shape of a physical resonating cavity, you can change the

419
00:24:11,119 --> 00:24:13,280
Speaker 3: sound of the vowel produced.

420
00:24:13,720 --> 00:24:16,960
Speaker 1: Right right. Another take on this. I was reading the

421
00:24:17,000 --> 00:24:20,160
Speaker 1: BBC Future article The Machines That Learned to Listen by

422
00:24:20,240 --> 00:24:24,960
Speaker 1: Kadia Musfych, and it describes these as resonance tubes connected

423
00:24:24,960 --> 00:24:28,320
Speaker 1: to organ pipes. So you know, this is not to

424
00:24:28,359 --> 00:24:29,960
Speaker 1: say that we have this is not on this like

425
00:24:30,000 --> 00:24:32,560
Speaker 1: the same level as some sort of imaginary brazen head

426
00:24:32,600 --> 00:24:34,680
Speaker 1: that's going to speak of its own and spout out,

427
00:24:34,960 --> 00:24:37,679
Speaker 1: spit out wisdom for you to interpret. This is about

428
00:24:37,720 --> 00:24:41,320
Speaker 1: just figuring out, you know, how these vowel sounds are

429
00:24:41,440 --> 00:24:47,400
Speaker 1: produced and reproducing them through a basic mechanical system. Musvich

430
00:24:47,480 --> 00:24:49,560
Speaker 1: also points out a few other key individuals in the

431
00:24:49,600 --> 00:24:54,439
Speaker 1: advancement of this technology. There's Wolfgang von Kimplin in Vienna,

432
00:24:54,480 --> 00:24:57,640
Speaker 1: who created a similar acoustic mechanical speech machine about ten

433
00:24:57,720 --> 00:25:02,880
Speaker 1: years after Kratzenstein. And then she also mentions English inventor

434
00:25:03,160 --> 00:25:06,120
Speaker 1: Charles Wheatstone, who would improve on this in the early

435
00:25:06,200 --> 00:25:07,040
Speaker 1: nineteenth century.

436
00:25:07,440 --> 00:25:10,159
Speaker 3: Charles Wheatstone. I'm going to mention him again in a minute,

437
00:25:10,200 --> 00:25:13,000
Speaker 3: but he's also notable because he was one of the

438
00:25:13,040 --> 00:25:18,439
Speaker 3: inventors of the first commercially successful form of the telegraph.

439
00:25:18,920 --> 00:25:21,040
Speaker 3: So we talked about him in our episode on your

440
00:25:21,200 --> 00:25:24,399
Speaker 3: mention of the telegraph. But when it comes to the

441
00:25:24,400 --> 00:25:27,199
Speaker 3: one you mentioned before, that von Kemplan's machine, this is

442
00:25:27,240 --> 00:25:34,560
Speaker 3: interesting because I read that while this machine was allegedly real,

443
00:25:34,640 --> 00:25:37,119
Speaker 3: it was a real attempt to make a machine that

444
00:25:37,160 --> 00:25:41,720
Speaker 3: would speak. Von Kemplan is now known for essentially being

445
00:25:41,760 --> 00:25:45,320
Speaker 3: a hoaxer because he tried to create other automata, including

446
00:25:45,720 --> 00:25:49,280
Speaker 3: a chess playing automaton that was actually a hoax. It

447
00:25:49,320 --> 00:25:52,080
Speaker 3: had a human inside it doing the moves, so it

448
00:25:52,119 --> 00:25:53,640
Speaker 3: was a fake robot.

449
00:25:53,680 --> 00:25:56,920
Speaker 1: Though as a fake still really impressive. It's interesting where

450
00:25:56,960 --> 00:25:59,000
Speaker 1: you get in like what sometimes you're wondering. You have

451
00:25:59,040 --> 00:26:02,040
Speaker 1: to wonder what the line knows between, you know, the

452
00:26:02,080 --> 00:26:06,040
Speaker 1: actual technological innovation and trickery. I mean, obviously it's deception,

453
00:26:06,400 --> 00:26:09,160
Speaker 1: and if you have a secret chamber in which there's

454
00:26:09,200 --> 00:26:11,800
Speaker 1: a whole person doing stuff, you know, that's a real

455
00:26:11,840 --> 00:26:15,119
Speaker 1: red flag there as well. But still the trickery is

456
00:26:15,119 --> 00:26:16,160
Speaker 1: pretty ingenious too.

457
00:26:16,640 --> 00:26:19,879
Speaker 3: Yeah, well, yeah, I mean it takes skill to be

458
00:26:19,880 --> 00:26:20,600
Speaker 3: a good magician.

459
00:26:20,960 --> 00:26:21,440
Speaker 1: Yeah.

460
00:26:21,480 --> 00:26:24,159
Speaker 3: Anyway, this brings us to the example that I was

461
00:26:24,200 --> 00:26:27,439
Speaker 3: really excited to talk about in today's episode, which is

462
00:26:27,480 --> 00:26:33,520
Speaker 3: the speaking machine of a nineteenth century inventor named Joseph Fober.

463
00:26:34,320 --> 00:26:37,160
Speaker 3: So one of my main sources here is just generally

464
00:26:37,160 --> 00:26:40,080
Speaker 3: a good source on the history of speech synthesis and

465
00:26:40,280 --> 00:26:44,639
Speaker 3: talking machines. It was a book chapter in the Rutledge

466
00:26:44,640 --> 00:26:48,480
Speaker 3: Handbook of Phonetics from twenty nineteen by an author named

467
00:26:48,520 --> 00:26:52,159
Speaker 3: Brad H. Story, who is part of the faculty of

468
00:26:52,200 --> 00:26:54,840
Speaker 3: the Department of Speech, Language and Hearing Sciences at the

469
00:26:54,920 --> 00:27:00,199
Speaker 3: University of Arizona. And Story, in this chapter traces the

470
00:27:00,280 --> 00:27:03,679
Speaker 3: history of speech synthesis from the mechanical methods of the

471
00:27:03,720 --> 00:27:07,560
Speaker 3: eighteenth and nineteenth centuries to the digital techniques of the present.

472
00:27:07,600 --> 00:27:10,960
Speaker 3: So it's the whole sort of modern arc of these machines.

473
00:27:11,359 --> 00:27:13,000
Speaker 3: But the thing I really want to focus in on

474
00:27:13,040 --> 00:27:15,960
Speaker 3: here now is this machine that I mentioned a minute ago,

475
00:27:16,000 --> 00:27:21,200
Speaker 3: by the nineteenth century German inventor Joseph Fober. This features

476
00:27:21,200 --> 00:27:24,639
Speaker 3: heavily at the beginning of stories chapter here. So this

477
00:27:24,800 --> 00:27:29,920
Speaker 3: machine was at various different times called the Marvelous Talking Machine.

478
00:27:30,119 --> 00:27:33,320
Speaker 3: You got a hyphen between talking machine and also the

479
00:27:33,600 --> 00:27:38,040
Speaker 3: euphonia from the Greek meaning good sound or sweet sound.

480
00:27:38,960 --> 00:27:40,960
Speaker 3: We'll see about that as we as we go on.

481
00:27:41,400 --> 00:27:44,800
Speaker 3: Robi included one illustration of the machine for you to

482
00:27:44,840 --> 00:27:46,640
Speaker 3: look at here. I think this may have been from

483
00:27:46,680 --> 00:27:49,760
Speaker 3: some kind of promotional material when this machine was featured

484
00:27:49,800 --> 00:27:51,720
Speaker 3: in an exhibit that I'll describe in a bit.

485
00:27:52,080 --> 00:27:55,439
Speaker 1: I love it in part because right there is this

486
00:27:55,600 --> 00:28:01,440
Speaker 1: angelic human face like right there on the machine, seemingly

487
00:28:01,560 --> 00:28:06,040
Speaker 1: as decoration or maybe tribute. I'm not sure, but I'm

488
00:28:06,040 --> 00:28:08,879
Speaker 1: not sure if it's actually necessary to the mechanics of

489
00:28:08,920 --> 00:28:09,320
Speaker 1: the device.

490
00:28:09,359 --> 00:28:13,359
Speaker 3: Here, I think it sort of is. I'll explain so.

491
00:28:13,600 --> 00:28:19,160
Speaker 3: Story introduces Fober's machine through the eyes of another inventor

492
00:28:19,200 --> 00:28:23,120
Speaker 3: and scientist of the day named Joseph Henry. A different Joseph,

493
00:28:23,520 --> 00:28:28,879
Speaker 3: a researcher on electromagnetic induction and also the inaugural secretary

494
00:28:28,920 --> 00:28:34,840
Speaker 3: of the Smithsonian Institution. Henry encountered Fober's Marvelous Talking Machine

495
00:28:34,880 --> 00:28:39,480
Speaker 3: at a private exhibition in Philadelphia on December twentieth, eighteen

496
00:28:39,640 --> 00:28:44,000
Speaker 3: forty five, and he described the demonstration in a letter

497
00:28:44,160 --> 00:28:46,959
Speaker 3: to a colleague named H. M. Alexander. So we have

498
00:28:47,080 --> 00:28:50,240
Speaker 3: contemporaneous notes on what it was doing and what it

499
00:28:50,240 --> 00:28:54,080
Speaker 3: looked like in this private demonstration. So here's how it worked.

500
00:28:54,760 --> 00:28:59,720
Speaker 3: It was controlled by an operator via a mainly by

501
00:29:00,040 --> 00:29:03,880
Speaker 3: foot pedals and a keyboard, essentially just like an organ,

502
00:29:04,040 --> 00:29:07,120
Speaker 3: like a chamber organ, and in fact the device could

503
00:29:07,160 --> 00:29:10,800
Speaker 3: in some ways be considered a modified organ. So you

504
00:29:10,840 --> 00:29:14,120
Speaker 3: had a foot pedal that operated a bellows and that

505
00:29:14,120 --> 00:29:18,480
Speaker 3: would supply airflow to the whole system, and the bellows

506
00:29:18,560 --> 00:29:22,560
Speaker 3: pumped air through an artificial larynx that had vocal cords

507
00:29:22,600 --> 00:29:24,440
Speaker 3: that were in this source said to be made of

508
00:29:24,560 --> 00:29:30,520
Speaker 3: rubber and these so this artificial glottis or artificial vocal

509
00:29:30,560 --> 00:29:35,120
Speaker 3: cords would vibrate to produce the fundamental sound of the

510
00:29:35,160 --> 00:29:38,280
Speaker 3: machine's voice when air was flowing through them. And then

511
00:29:38,320 --> 00:29:42,520
Speaker 3: you had sixteen keys on the keyboard which were connected

512
00:29:42,560 --> 00:29:47,160
Speaker 3: by strings and levers to the various components that controlled

513
00:29:47,320 --> 00:29:50,120
Speaker 3: the shaping of that sound of that, you know, the

514
00:29:50,160 --> 00:29:54,000
Speaker 3: resonating sound from that airflow through the glottis into speech.

515
00:29:54,480 --> 00:29:57,040
Speaker 3: One of the interesting things is, as we've been saying,

516
00:29:57,040 --> 00:29:59,920
Speaker 3: this device actually had a face, so the face was

517
00:30:00,120 --> 00:30:03,920
Speaker 3: made of carved wood, essentially a large doll head, but

518
00:30:04,040 --> 00:30:06,800
Speaker 3: it had a hinged jaw, so maybe you should think

519
00:30:06,840 --> 00:30:09,640
Speaker 3: of it more like a ventriloquist dummy. You're loving this,

520
00:30:09,720 --> 00:30:13,560
Speaker 3: aren't you, Yeah, Night of the Living dummy. But it

521
00:30:13,600 --> 00:30:18,000
Speaker 3: can actually speak, And so inside the dummy's mouth there

522
00:30:18,080 --> 00:30:22,360
Speaker 3: was an ivory tongue that could be moved around inside

523
00:30:22,400 --> 00:30:26,640
Speaker 3: the oral cavity to control the shape of the resonating chamber.

524
00:30:27,360 --> 00:30:30,360
Speaker 3: And by controlling these different elements like the mouth and

525
00:30:30,400 --> 00:30:33,400
Speaker 3: the tongue and all that with the keys on the keyboard,

526
00:30:34,600 --> 00:30:38,720
Speaker 3: it quote imposed time varying changes to the air cavity

527
00:30:39,040 --> 00:30:45,760
Speaker 3: appropriate for generating apparently convincing renditions of connected speech. So

528
00:30:45,800 --> 00:30:48,960
Speaker 3: it may not have sounded perfect or even pleasant, but

529
00:30:49,240 --> 00:30:53,080
Speaker 3: apparently people in the room could understand what the machine

530
00:30:53,120 --> 00:30:55,200
Speaker 3: was saying when Fober operated it.

531
00:30:55,640 --> 00:30:56,160
Speaker 1: So this is.

532
00:30:56,080 --> 00:30:59,520
Speaker 3: Eighteen forty five and the machine is speaking intelligible words.

533
00:31:00,360 --> 00:31:03,800
Speaker 3: Henry in this letter compares it favorably to a different

534
00:31:03,840 --> 00:31:06,600
Speaker 3: talking machine, one he had seen years before. This was

535
00:31:06,640 --> 00:31:08,760
Speaker 3: one of the ones you mentioned, Rob, the one built

536
00:31:08,800 --> 00:31:12,200
Speaker 3: by the English scientist and inventor Charles Wheatstone. Again, the

537
00:31:12,520 --> 00:31:18,760
Speaker 3: telegraph guy Wheatstone's talking machine was capable of being understood

538
00:31:18,920 --> 00:31:22,240
Speaker 3: for the set of words it could produce, but Fober's

539
00:31:22,280 --> 00:31:27,400
Speaker 3: machine was far superior because its speech repertoire was infinitely variable,

540
00:31:27,440 --> 00:31:31,280
Speaker 3: so he could speak whole sentences, and those sentences could

541
00:31:31,320 --> 00:31:34,360
Speaker 3: contain any words and any sounds you wanted, as long

542
00:31:34,400 --> 00:31:37,440
Speaker 3: as they were in one of the covered languages. Obviously

543
00:31:37,480 --> 00:31:40,840
Speaker 3: it couldn't do, you know, like tonal languages or like

544
00:31:41,000 --> 00:31:44,120
Speaker 3: speak Mandarin or something, but it seems like mainly it

545
00:31:44,160 --> 00:31:46,520
Speaker 3: was speaking German and English. It was said at the

546
00:31:46,520 --> 00:31:49,400
Speaker 3: time that it could speak any European language. Now, I

547
00:31:49,440 --> 00:31:51,680
Speaker 3: think one thing that's really worth noting here is that

548
00:31:51,800 --> 00:31:54,240
Speaker 3: if you imagine how a machine like this would work,

549
00:31:54,800 --> 00:31:59,880
Speaker 3: the success of the performance would depend heavily on the

550
00:32:00,240 --> 00:32:04,120
Speaker 3: skill of the operator, since the speech patterns are not

551
00:32:04,320 --> 00:32:09,160
Speaker 3: like programmed, and you know, it's not sort of expressed automatically,

552
00:32:09,720 --> 00:32:13,760
Speaker 3: but expressed in real time by the player operating the

553
00:32:13,760 --> 00:32:16,000
Speaker 3: bellows and the keys. And I think also there were

554
00:32:16,040 --> 00:32:19,640
Speaker 3: some screws and stuff that would manipulate pitch and things

555
00:32:19,680 --> 00:32:22,280
Speaker 3: like that, So you have to play this just like

556
00:32:22,360 --> 00:32:26,600
Speaker 3: you would play a musical instrument. So different players using

557
00:32:26,640 --> 00:32:30,840
Speaker 3: the same machine would probably produce fairly different sounding speech,

558
00:32:31,240 --> 00:32:34,200
Speaker 3: even if they had memorized which keys corresponded to which

559
00:32:34,240 --> 00:32:38,600
Speaker 3: phonetic units. So nobody I've read says this, but you know,

560
00:32:38,760 --> 00:32:42,120
Speaker 3: I'm kind of picturing Fober as a sort of phantom

561
00:32:42,160 --> 00:32:44,480
Speaker 3: of the opera at the at the organ keyboard. You know,

562
00:32:44,520 --> 00:32:46,440
Speaker 3: he's not just like pressing the keys, but giving a

563
00:32:46,480 --> 00:32:51,000
Speaker 3: real passionate and dramatic performance when somebody sells it. Ye yeah,

564
00:32:51,120 --> 00:32:54,680
Speaker 3: make it say Como tale vu or whatever. It also

565
00:32:54,720 --> 00:32:56,760
Speaker 3: sang songs, by the way. I'll get into that in

566
00:32:56,800 --> 00:32:59,120
Speaker 3: a minute. But I was wondering, what what did what

567
00:32:59,400 --> 00:33:01,840
Speaker 3: did people asking, you know, what's the equivalent in eighteen

568
00:33:02,000 --> 00:33:04,720
Speaker 3: forty five of yelling out, you know, play Freebird? And

569
00:33:04,760 --> 00:33:07,080
Speaker 3: I was thinking, maybe it's people are yelling for Tipicanu

570
00:33:07,120 --> 00:33:07,840
Speaker 3: and Tyler too.

571
00:33:08,400 --> 00:33:09,880
Speaker 1: Oh yeah.

572
00:33:10,040 --> 00:33:12,960
Speaker 3: So an interesting detail that story includes in this chapter

573
00:33:13,160 --> 00:33:16,200
Speaker 3: is that this was not the first time Fober had

574
00:33:16,240 --> 00:33:19,080
Speaker 3: built a talking machine. In fact, this was not the

575
00:33:19,080 --> 00:33:23,040
Speaker 3: first time Fober had built this exact talking machine. There

576
00:33:23,160 --> 00:33:26,040
Speaker 3: was an earlier version of it that was destroyed by

577
00:33:26,160 --> 00:33:30,080
Speaker 3: Fober himself. Quote in a bout of depression and intoxication.

578
00:33:30,960 --> 00:33:33,400
Speaker 3: I should say that nearly every source I read on

579
00:33:33,480 --> 00:33:38,200
Speaker 3: Fober mentions something about him being disheveled or even haunted,

580
00:33:38,840 --> 00:33:42,719
Speaker 3: obsessed with his machine, and generally emotionally unwell or at

581
00:33:42,720 --> 00:33:45,240
Speaker 3: the very least having a really rough time a lot

582
00:33:45,240 --> 00:33:49,280
Speaker 3: of the time. Multiple writers describe him in terms containing

583
00:33:49,280 --> 00:33:54,080
Speaker 3: a lot of pity. But so, it took Fober apparently

584
00:33:54,160 --> 00:33:57,880
Speaker 3: twenty years to perfect the first version of the machine,

585
00:33:57,920 --> 00:34:01,400
Speaker 3: the one that he drunkenly destroyed, but he was able

586
00:34:01,480 --> 00:34:04,480
Speaker 3: to recreate the second version within a year of that.

587
00:34:05,000 --> 00:34:07,360
Speaker 3: And this kind of suggests to me the possibility that

588
00:34:07,960 --> 00:34:11,560
Speaker 3: the original creation of the machine may have really been

589
00:34:11,880 --> 00:34:15,600
Speaker 3: a project of fundamental research about phonetics more than it

590
00:34:15,719 --> 00:34:19,280
Speaker 3: was about engineering. And so once he had the knowledge

591
00:34:19,280 --> 00:34:22,080
Speaker 3: in hand of how each sound was produced, like what

592
00:34:22,160 --> 00:34:25,000
Speaker 3: the shape of the oral cavity, you know, how that

593
00:34:25,040 --> 00:34:29,120
Speaker 3: corresponded to the sounds, recreating the machine itself might have

594
00:34:29,200 --> 00:34:32,080
Speaker 3: been a relatively simple proposition. Is really what you needed

595
00:34:32,160 --> 00:34:36,240
Speaker 3: was the knowledge about how phonetics correspond to physical shapes.

596
00:34:36,800 --> 00:34:38,840
Speaker 1: Yeah, And if he had that, and certainly if he

597
00:34:38,880 --> 00:34:42,400
Speaker 1: had notes on the matter and his designs recorded, it

598
00:34:42,440 --> 00:34:45,600
Speaker 1: would be easier to come back and reproduce that. Yeah.

599
00:34:46,360 --> 00:34:51,360
Speaker 3: So Joseph Henry's letter about Fober's talking machine demonstration. It

600
00:34:51,400 --> 00:34:55,320
Speaker 3: also includes speculation about the uses to which a machine

601
00:34:55,360 --> 00:34:59,080
Speaker 3: like this could be put. One interesting idea he has is,

602
00:34:59,280 --> 00:35:02,080
Speaker 3: what if you could take a spoken message at one

603
00:35:02,120 --> 00:35:07,080
Speaker 3: location and code that spoken message into inputs on this

604
00:35:07,239 --> 00:35:13,120
Speaker 3: keyboard on this machine, and then, through electromagnetic means, transmit

605
00:35:13,239 --> 00:35:17,920
Speaker 3: those keystrokes across wires to a totally separate second location,

606
00:35:18,680 --> 00:35:22,480
Speaker 3: and then those electrical signals could operate the speech organs

607
00:35:22,520 --> 00:35:26,040
Speaker 3: of the doll faced machine. In the second location. You

608
00:35:26,080 --> 00:35:32,480
Speaker 3: would essentially be transmitting speech itself across great distance. Notable

609
00:35:32,480 --> 00:35:36,080
Speaker 3: that Henry's idea here is roughly thirty years before Alexander

610
00:35:36,080 --> 00:35:39,560
Speaker 3: Graham Bell demonstrates the principle of the telephone. But there

611
00:35:39,600 --> 00:35:43,400
Speaker 3: is a very important difference, which is that while Bell's

612
00:35:43,440 --> 00:35:46,960
Speaker 3: telephone and these are stories words here quote transmitted an

613
00:35:47,000 --> 00:35:52,080
Speaker 3: electrical analog of the speech pressure wave. Henry's description alluded

614
00:35:52,120 --> 00:35:57,400
Speaker 3: to representing speech in compressed form based on slowly varying

615
00:35:57,480 --> 00:36:01,319
Speaker 3: movements of the operator's hands, fingers, and feet as they

616
00:36:01,360 --> 00:36:05,319
Speaker 3: formed the keystroke sequences required to produce an utterance, a

617
00:36:05,360 --> 00:36:09,120
Speaker 3: signal processing technique that would not be implemented into telephone

618
00:36:09,160 --> 00:36:13,600
Speaker 3: transmission systems for nearly another century. So the interesting thing

619
00:36:13,640 --> 00:36:17,279
Speaker 3: about Henry here is that he's not just imagining converting

620
00:36:17,320 --> 00:36:20,600
Speaker 3: the sound of a voice into an impulse that travels

621
00:36:20,640 --> 00:36:24,719
Speaker 3: along the wire. He's imagining a coding process. It's put

622
00:36:24,840 --> 00:36:28,160
Speaker 3: into code for the transmission and then decoded by the

623
00:36:28,200 --> 00:36:29,399
Speaker 3: machine at the other end.

624
00:36:30,080 --> 00:36:33,560
Speaker 1: I can't help but try to imagine this alternate past

625
00:36:33,840 --> 00:36:38,000
Speaker 1: in which instead of early telephones, people all had this

626
00:36:38,760 --> 00:36:43,560
Speaker 1: weird cherub head mounted on the wall that then speaks

627
00:36:43,600 --> 00:36:47,680
Speaker 1: to you in this I'm assumed slightly haunting voice. Oh.

628
00:36:47,760 --> 00:36:50,880
Speaker 3: I'll get to the haunting voice in a second, but anyway,

629
00:36:50,960 --> 00:36:54,520
Speaker 3: story flags it as historically significant that this one invention

630
00:36:54,680 --> 00:36:58,960
Speaker 3: had both succeeded in producing generally intelligible synthetic speech to

631
00:36:59,040 --> 00:37:02,279
Speaker 3: people in the room with it, and it had inspired

632
00:37:02,360 --> 00:37:05,480
Speaker 3: at least one onlooker to start considering ideas for the

633
00:37:05,560 --> 00:37:09,560
Speaker 3: electrical transmission of low bandwidth speech from one place to another.

634
00:37:10,480 --> 00:37:14,359
Speaker 3: But neither of these possibilities really went anywhere. Henry did

635
00:37:14,440 --> 00:37:18,239
Speaker 3: not devote any more effort to musing about the electrical transmission,

636
00:37:18,840 --> 00:37:22,880
Speaker 3: and Fober's machine ended up being a circus side show

637
00:37:23,040 --> 00:37:28,480
Speaker 3: almost literally. So after this, Fober needed money, and beginning

638
00:37:28,480 --> 00:37:31,880
Speaker 3: in eighteen forty six, to get money, he signed on

639
00:37:32,080 --> 00:37:35,279
Speaker 3: to demonstrate his machine for P. T. Barnum. Gotta have

640
00:37:35,320 --> 00:37:39,319
Speaker 3: something for everybody, even people who want a talking doll

641
00:37:39,400 --> 00:37:44,360
Speaker 3: head operated by a disheveled German organ master. So Fober

642
00:37:44,400 --> 00:37:48,759
Speaker 3: committed to exhibit the marvelous Speaking Machine for Barnum at

643
00:37:48,800 --> 00:37:52,400
Speaker 3: the Egyptian Hall in London. This was like a general

644
00:37:52,480 --> 00:37:56,280
Speaker 3: exhibition hall in Piccadilly which hosted all kinds of shows,

645
00:37:56,320 --> 00:37:58,680
Speaker 3: but I think, especially in the latter part of the

646
00:37:58,760 --> 00:38:01,680
Speaker 3: nineteenth century, it was known for showing like a lot

647
00:38:01,719 --> 00:38:07,440
Speaker 3: of Mountebanks and fraudulent spiritualist demonstrators. Yeah, I'll reveal to

648
00:38:07,480 --> 00:38:10,359
Speaker 3: you that you're actually a reincarnation of Cleopatra.

649
00:38:11,760 --> 00:38:12,200
Speaker 1: Lucky you.

650
00:38:12,840 --> 00:38:15,160
Speaker 3: But by noting that that's just a random thing, I'm

651
00:38:15,200 --> 00:38:18,279
Speaker 3: not trying to cast dispersions on Fober because I want

652
00:38:18,320 --> 00:38:21,480
Speaker 3: to stress that it seems totally clear that Fober was

653
00:38:21,640 --> 00:38:24,680
Speaker 3: no con artist. As best we can tell, his machine

654
00:38:24,760 --> 00:38:29,160
Speaker 3: really did work, and when played correctly, it did really

655
00:38:29,200 --> 00:38:34,080
Speaker 3: speak original sentences that people could, for the most part understand. Though,

656
00:38:34,200 --> 00:38:38,200
Speaker 3: one thing that emerges from reading descriptions of this is

657
00:38:38,239 --> 00:38:44,520
Speaker 3: that coding intelligible information and sounding like speech are two

658
00:38:44,560 --> 00:38:48,080
Speaker 3: completely different things. So it seems that a lot of

659
00:38:48,080 --> 00:38:51,880
Speaker 3: people could tell what the machine was saying, but still

660
00:38:52,000 --> 00:38:55,920
Speaker 3: they were not very impressed by what they heard. And

661
00:38:55,960 --> 00:39:00,840
Speaker 3: I found a spectacularly evocative description of what the machine

662
00:39:00,880 --> 00:39:04,000
Speaker 3: was like are recorded in a book called Instruments and

663
00:39:04,040 --> 00:39:08,040
Speaker 3: the Imagination by Thomas L. Hankins and Robert J. Silverman,

664
00:39:08,200 --> 00:39:11,759
Speaker 3: Princeton University Press, nineteen ninety nine. But the main thing

665
00:39:11,800 --> 00:39:14,759
Speaker 3: here is that they're quoting a person who saw the

666
00:39:14,760 --> 00:39:17,640
Speaker 3: machine in person in eighteen forty six, I believe, and

667
00:39:17,880 --> 00:39:20,400
Speaker 3: then wrote about it in a memoir. But generally the

668
00:39:20,440 --> 00:39:22,960
Speaker 3: authors here they note that there were like some satirical

669
00:39:23,080 --> 00:39:27,080
Speaker 3: articles making reference to Faber's machine, suggesting, for example, that

670
00:39:27,200 --> 00:39:29,759
Speaker 3: it could be used to replace the speaker of the

671
00:39:29,800 --> 00:39:34,759
Speaker 3: House of Commons. Yuk yah, those wacky politicians. But then

672
00:39:34,800 --> 00:39:37,160
Speaker 3: they well, they do kind of make a funny point. Actually,

673
00:39:37,239 --> 00:39:39,279
Speaker 3: they say, like you could just program it to say

674
00:39:39,440 --> 00:39:41,520
Speaker 3: order order at ten minute intervals.

675
00:39:43,480 --> 00:39:45,200
Speaker 1: Well that's pretty good, that's funny today.

676
00:39:45,600 --> 00:39:48,399
Speaker 3: Yeah. But anyway, then there's a part of the book

677
00:39:48,400 --> 00:39:52,359
Speaker 3: where they're including this evocative written account which is from

678
00:39:52,520 --> 00:39:57,560
Speaker 3: a London theater manager named John Hollingshead who saw this

679
00:39:57,640 --> 00:40:00,400
Speaker 3: machine in person when he was nineteen years old and

680
00:40:00,440 --> 00:40:03,239
Speaker 3: then wrote about it in a memoirs or some book.

681
00:40:03,520 --> 00:40:08,200
Speaker 3: But anyway, this is hallings Head's account. The exhibitor, Professor Fober,

682
00:40:08,640 --> 00:40:12,280
Speaker 3: was a sad faced man, dressed in respectable, well worn

683
00:40:12,320 --> 00:40:16,440
Speaker 3: clothes that were soiled by contact with tools, wood and machinery.

684
00:40:17,040 --> 00:40:20,600
Speaker 3: The room looked like a laboratory and workshop, which it was.

685
00:40:21,200 --> 00:40:23,960
Speaker 3: The professor was not too clean, and his hair and

686
00:40:24,040 --> 00:40:27,680
Speaker 3: beard sadly wanted the attention of a barber. I have

687
00:40:27,760 --> 00:40:30,439
Speaker 3: no doubt that he slept in the same room as

688
00:40:30,440 --> 00:40:33,720
Speaker 3: his figure, his scientific Frankenstein Monster.

689
00:40:34,320 --> 00:40:34,520
Speaker 1: Note.

690
00:40:34,520 --> 00:40:36,319
Speaker 3: I guess the novel would have only been a few

691
00:40:36,320 --> 00:40:38,000
Speaker 3: decades old at this time.

692
00:40:38,400 --> 00:40:40,879
Speaker 1: Yeah, yeah, eighteen eighteen on Frankenstein there.

693
00:40:41,040 --> 00:40:44,680
Speaker 3: Yeah, sorry going on with Halling's head, and I felt

694
00:40:44,680 --> 00:40:47,399
Speaker 3: the secret influence of an idea that the two were

695
00:40:47,520 --> 00:40:49,600
Speaker 3: destined to live and die together.

696
00:40:50,160 --> 00:40:53,839
Speaker 1: Oh my god, this is those pretty strong words. Yes.

697
00:40:54,880 --> 00:40:58,320
Speaker 3: The professor, with a slight German accent, put his wonderful

698
00:40:58,360 --> 00:41:02,000
Speaker 3: toy in motion. He explained its action. It was not

699
00:41:02,200 --> 00:41:06,400
Speaker 3: necessary to prove the absence of deception one keyboard touched

700
00:41:06,440 --> 00:41:10,279
Speaker 3: by the professor, produced words which slowly and deliberately, in

701
00:41:10,360 --> 00:41:14,440
Speaker 3: a hoarse, sepulchral voice, came from the mouth of the figure,

702
00:41:14,680 --> 00:41:17,640
Speaker 3: as if from the depths of a tomb. It wanted

703
00:41:17,680 --> 00:41:21,160
Speaker 3: little imagination to make the very few visitors believe that

704
00:41:21,200 --> 00:41:25,480
Speaker 3: the figure contained an imprisoned human or half human being

705
00:41:26,200 --> 00:41:30,719
Speaker 3: bound to speak slowly when tormented by the unseen power outside.

706
00:41:31,480 --> 00:41:33,640
Speaker 3: No one thought for a moment that they were being

707
00:41:33,680 --> 00:41:37,520
Speaker 3: fooled by a second edition of the Invisible Girl fraud.

708
00:41:38,440 --> 00:41:41,240
Speaker 3: And by the way, the reference to the Invisible Girl fraud,

709
00:41:41,280 --> 00:41:44,120
Speaker 3: I believe is about the many fake machines and fake

710
00:41:44,160 --> 00:41:47,520
Speaker 3: automata that were actually worked by having a human hidden

711
00:41:47,520 --> 00:41:50,800
Speaker 3: inside operating it. But going on, so Holling said, says,

712
00:41:51,160 --> 00:41:53,759
Speaker 3: nobody thought that there was an invisible girl operating. This

713
00:41:53,880 --> 00:41:57,440
Speaker 3: as clear, this is real. He goes on. There were truth,

714
00:41:57,760 --> 00:42:01,200
Speaker 3: laborious invention, and good faith in every part of the

715
00:42:01,239 --> 00:42:05,239
Speaker 3: melancholy room. As a crowning display, the head sang a

716
00:42:05,280 --> 00:42:10,440
Speaker 3: sepulchral version of God Save the Queen, which suggested, inevitably,

717
00:42:10,680 --> 00:42:14,799
Speaker 3: God save the inventor. This extraordinary effect was achieved by

718
00:42:14,840 --> 00:42:18,319
Speaker 3: the professor working two keyboards, one for the words and

719
00:42:18,360 --> 00:42:22,239
Speaker 3: one for the music. Never probably before or since, has

720
00:42:22,280 --> 00:42:26,279
Speaker 3: the national anthem been so sung, sadder and wiser. I

721
00:42:26,680 --> 00:42:30,360
Speaker 3: and the few visitors crept slowly from the place, leaving

722
00:42:30,400 --> 00:42:33,560
Speaker 3: the Professor with his one and only treasure, his child

723
00:42:33,640 --> 00:42:36,320
Speaker 3: of infinite labor and unmeasurable sorrow.

724
00:42:36,760 --> 00:42:41,279
Speaker 1: Oh wow, that is a lot. I mean, obviously he

725
00:42:41,680 --> 00:42:44,160
Speaker 1: lays it on really thick about the sadness of the

726
00:42:44,160 --> 00:42:47,120
Speaker 1: inventor here. And then also there's the ideas like this

727
00:42:47,239 --> 00:42:49,960
Speaker 1: was no hoax, this was real and it was depressing.

728
00:42:50,360 --> 00:42:54,840
Speaker 3: Yeah, it's a weird mix of like like pity but

729
00:42:55,080 --> 00:42:59,319
Speaker 3: real admiration, you know that, Like, there's something beautiful and

730
00:42:59,440 --> 00:43:02,799
Speaker 3: honest and true about this machine and his devotion to

731
00:43:02,840 --> 00:43:05,520
Speaker 3: it and the genius it took to create it. But

732
00:43:05,640 --> 00:43:09,600
Speaker 3: also it makes everybody feel bad and nobody wants to

733
00:43:09,600 --> 00:43:11,880
Speaker 3: look at it or listen to it, and everybody leaves

734
00:43:11,880 --> 00:43:16,400
Speaker 3: feeling depressed. YEA, something about that struck me as actually

735
00:43:16,760 --> 00:43:19,400
Speaker 3: quite poignant and meaningful. Maybe we can come back to

736
00:43:19,440 --> 00:43:21,600
Speaker 3: that in a minute, but I did want to flag

737
00:43:21,640 --> 00:43:25,040
Speaker 3: that there was one notable visitor who, coming back to

738
00:43:25,040 --> 00:43:29,080
Speaker 3: the Invisible Girl suspicion, he did at first suspect fraud,

739
00:43:29,200 --> 00:43:32,359
Speaker 3: and that was the Duke of Wellington. I was reading

740
00:43:32,360 --> 00:43:34,680
Speaker 3: about this in a book called The Shows of London

741
00:43:34,760 --> 00:43:39,440
Speaker 3: by Richard Daniel Atlick, and at Lick recounts that Wellington,

742
00:43:39,760 --> 00:43:42,719
Speaker 3: when he first went to the demonstration, he was so

743
00:43:42,880 --> 00:43:47,160
Speaker 3: impressed by Faber's speaking machine that he asked to be

744
00:43:47,239 --> 00:43:50,000
Speaker 3: allowed to touch the keys with his own fingers, you know,

745
00:43:50,080 --> 00:43:52,799
Speaker 3: so he could see that it was genuine. And then

746
00:43:52,920 --> 00:43:55,239
Speaker 3: he did confirm that it was genuine, and then he

747
00:43:55,320 --> 00:43:58,239
Speaker 3: insisted that he'd be taught how to use it. So

748
00:43:58,480 --> 00:44:01,520
Speaker 3: Fober taught the Duke to play the machine in both

749
00:44:01,600 --> 00:44:05,560
Speaker 3: German and English, and Wellington did get it like he could.

750
00:44:05,600 --> 00:44:08,080
Speaker 3: He could make it speak sentences in German and English,

751
00:44:08,200 --> 00:44:11,120
Speaker 3: and he was amazed, writing in the visitor's log of

752
00:44:11,200 --> 00:44:15,239
Speaker 3: the exhibit that the speaking machine, or the Euphonia, was

753
00:44:15,360 --> 00:44:30,040
Speaker 3: quote an extraordinary production of mechanical genius. Faber's machine also

754
00:44:30,080 --> 00:44:33,840
Speaker 3: got rave reviews in The Times, in the Illustrated London News.

755
00:44:33,880 --> 00:44:37,440
Speaker 3: A lot of people like looked at it and they

756
00:44:37,920 --> 00:44:40,440
Speaker 3: thought that, like, yeah, this is a work of genius.

757
00:44:40,280 --> 00:44:43,959
Speaker 3: It's incredible that he's done this. But at the same time,

758
00:44:44,560 --> 00:44:48,680
Speaker 3: audiences really were not into it. Barnum himself noticed that

759
00:44:48,719 --> 00:44:52,040
Speaker 3: Fober's machine was not attracting crowds, it was not selling

760
00:44:52,080 --> 00:44:56,520
Speaker 3: tickets and not generating revenue, and so eventually he took

761
00:44:56,600 --> 00:45:00,360
Speaker 3: Fober's machine out of the Egyptian Hall in Life, London

762
00:45:00,880 --> 00:45:03,680
Speaker 3: and added it to a traveling exhibit that went around

763
00:45:03,719 --> 00:45:08,480
Speaker 3: the English countryside doing performances. And from here Faber himself

764
00:45:08,520 --> 00:45:11,759
Speaker 3: seems to kind of disappear from the historical record. Some

765
00:45:11,840 --> 00:45:15,439
Speaker 3: sources indicate that he may have died by suicide during

766
00:45:15,480 --> 00:45:18,880
Speaker 3: this period, though that isn't known for sure. But after

767
00:45:19,320 --> 00:45:23,480
Speaker 3: historical sources stopped mentioning Faber himself, they still make references

768
00:45:23,520 --> 00:45:27,600
Speaker 3: to his machine, reading from story here quote. Although his

769
00:45:27,760 --> 00:45:31,440
Speaker 3: talking machine continued to make side show like appearances in

770
00:45:31,480 --> 00:45:35,000
Speaker 3: Europe and North America over the next thirty years. It

771
00:45:35,080 --> 00:45:38,239
Speaker 3: seems a relative, perhaps a niece or nephew, may have

772
00:45:38,320 --> 00:45:41,600
Speaker 3: inherited the machine and performed with it to generate income.

773
00:45:42,600 --> 00:45:45,800
Speaker 1: So maybe, no matter whatever happened to him, maybe a

774
00:45:45,840 --> 00:45:49,160
Speaker 1: relative with a little more showmanship like stepped in and

775
00:45:49,360 --> 00:45:51,640
Speaker 1: was able to make at least some sort of an

776
00:45:51,680 --> 00:45:52,440
Speaker 1: income off of it.

777
00:45:53,000 --> 00:45:55,680
Speaker 3: Yes, But then again, like I'm struck by the strange

778
00:45:55,719 --> 00:46:02,400
Speaker 3: ironic sadness of this, this was actually a scientifically significant invention,

779
00:46:02,719 --> 00:46:06,440
Speaker 3: like he had done something kind of amazing, but it

780
00:46:06,560 --> 00:46:10,040
Speaker 3: just never really went anywhere under his mastery. And then yeah,

781
00:46:10,120 --> 00:46:15,280
Speaker 3: maybe a relative was a better Carnival Barker essentially to

782
00:46:15,320 --> 00:46:17,759
Speaker 3: perform with the machine and make some money off of it.

783
00:46:17,920 --> 00:46:20,239
Speaker 1: I mean, it reminds me of so many advancements in

784
00:46:20,440 --> 00:46:25,960
Speaker 1: say robotics that we've seen over the years, where oftentimes,

785
00:46:26,160 --> 00:46:28,920
Speaker 1: you know, to a certain extent, unfairly, they'll just be

786
00:46:29,000 --> 00:46:31,560
Speaker 1: one little clip of it that goes viral and people

787
00:46:31,600 --> 00:46:35,880
Speaker 1: react to be it some sort of you know, human

788
00:46:36,000 --> 00:46:39,080
Speaker 1: likeness with facial features that seem to be moving or

789
00:46:39,120 --> 00:46:42,440
Speaker 1: operating in an uncanny way, or something like the various

790
00:46:43,239 --> 00:46:48,680
Speaker 1: dog robots from Boston Dynamics that are very impressive but

791
00:46:48,760 --> 00:46:52,560
Speaker 1: also maybe interpreted as being a bit creepy. And so

792
00:46:52,760 --> 00:46:54,680
Speaker 1: even though they are these, you know, they are often

793
00:46:54,920 --> 00:47:00,440
Speaker 1: examples of a real impressive technological advancement. Setting a side

794
00:47:00,760 --> 00:47:04,160
Speaker 1: actual applications, you can have a situation where something like

795
00:47:04,200 --> 00:47:10,359
Speaker 1: that is not as comforting, not as entertaining as say

796
00:47:10,360 --> 00:47:13,360
Speaker 1: an act of puppetry or even an act of just

797
00:47:13,719 --> 00:47:17,520
Speaker 1: outright well, maybe not fraud, but say a robot or

798
00:47:17,560 --> 00:47:23,480
Speaker 1: a costume depicting a robot maybe ultimately maybe more reassuring,

799
00:47:23,560 --> 00:47:25,799
Speaker 1: maybe more fun compared to the actual thing.

800
00:47:26,360 --> 00:47:28,680
Speaker 3: Well yeah, which may which may just be fun or

801
00:47:28,719 --> 00:47:31,520
Speaker 3: may in fact be fraud, depending on what exactly they're

802
00:47:31,520 --> 00:47:34,840
Speaker 3: saying about it. Yeah, but this is a great point

803
00:47:34,880 --> 00:47:36,680
Speaker 3: and it brings me to I just wanted to mention

804
00:47:36,719 --> 00:47:40,120
Speaker 3: a few of the the general notes about the history

805
00:47:40,160 --> 00:47:43,959
Speaker 3: of speech synthesis from the end of this this book

806
00:47:44,040 --> 00:47:47,719
Speaker 3: chapter by Brad's story Story writs that, you know, while

807
00:47:47,719 --> 00:47:52,200
Speaker 3: there are technological use cases for speech synthesizers, we've you know,

808
00:47:52,239 --> 00:47:55,600
Speaker 3: we've got a number of them operating in consumer technology today,

809
00:47:56,520 --> 00:48:00,360
Speaker 3: and even before you had you know, personal digitalist stunts

810
00:48:00,400 --> 00:48:03,680
Speaker 3: and stuff, there would be use cases for speech synthesizers,

811
00:48:04,239 --> 00:48:07,440
Speaker 3: for example, people who have a disability that makes it

812
00:48:07,480 --> 00:48:10,960
Speaker 3: difficult or impossible for them to speak. Another one is

813
00:48:11,000 --> 00:48:13,880
Speaker 3: that apparently this was actually used by the Allies in

814
00:48:13,920 --> 00:48:16,480
Speaker 3: World War Two. There were some forms of speech synthesis

815
00:48:16,480 --> 00:48:22,600
Speaker 3: that would allow sort of covert coded transmissions of something

816
00:48:22,800 --> 00:48:24,480
Speaker 3: like a phone call, So you could have a phone

817
00:48:24,480 --> 00:48:28,040
Speaker 3: call between like FDR and Winston Churchill. It's not really

818
00:48:28,040 --> 00:48:32,400
Speaker 3: a phone call. It's like a transmitted synthesized bit of speech,

819
00:48:32,520 --> 00:48:35,600
Speaker 3: and so it's very secure, but it doesn't sound like

820
00:48:35,640 --> 00:48:38,560
Speaker 3: the person talking. It sounds maybe more like the euphonia,

821
00:48:38,920 --> 00:48:43,320
Speaker 3: kind of robotic and unnatural and maybe making the president's

822
00:48:43,560 --> 00:48:47,520
Speaker 3: giggle a bit a president Prime minister. But anyway, So

823
00:48:48,000 --> 00:48:50,439
Speaker 3: what story says is that a large number of these

824
00:48:50,480 --> 00:48:55,040
Speaker 3: systems have actually been primarily used as research tools, as

825
00:48:55,080 --> 00:49:00,520
Speaker 3: scientific tools for understanding the nature of human speech. I

826
00:49:00,600 --> 00:49:05,000
Speaker 3: trying to reproduce human speech and failing at it, that

827
00:49:05,080 --> 00:49:09,040
Speaker 3: we come closer to understanding how speech actually works in

828
00:49:09,560 --> 00:49:12,600
Speaker 3: the human body. But the second general observation that I

829
00:49:12,600 --> 00:49:15,239
Speaker 3: thought is interesting, and this seems to be very much

830
00:49:15,280 --> 00:49:20,280
Speaker 3: reflected in the Fober's machine example. It is much easier

831
00:49:20,320 --> 00:49:24,359
Speaker 3: to create a machine that can speak intelligibly than one

832
00:49:24,400 --> 00:49:29,440
Speaker 3: that can speak naturally. So that indicates that when we talk,

833
00:49:29,600 --> 00:49:32,640
Speaker 3: there's actually more than one thing going on. Yes, we

834
00:49:32,800 --> 00:49:37,960
Speaker 3: are conveying mental information coded in words, and the substance

835
00:49:38,000 --> 00:49:41,360
Speaker 3: of that coding is phonetic. It's a series of sounds.

836
00:49:41,400 --> 00:49:44,879
Speaker 3: But of course, you know, the ironic thing to people

837
00:49:44,880 --> 00:49:47,120
Speaker 3: who were used to thinking about words as text is

838
00:49:47,120 --> 00:49:50,680
Speaker 3: that the phonetic core of language long predates writing, so

839
00:49:50,800 --> 00:49:53,760
Speaker 3: like the written text of a word is a visual

840
00:49:53,840 --> 00:49:56,400
Speaker 3: code for the sound of the word, which is the

841
00:49:56,440 --> 00:50:00,120
Speaker 3: code for its meaning. But anyway, so machines for hundred

842
00:50:00,200 --> 00:50:03,720
Speaker 3: of yours have been able to produce more or less

843
00:50:04,040 --> 00:50:07,399
Speaker 3: intelligible phonetic code. They can speak words, and people can

844
00:50:07,520 --> 00:50:11,520
Speaker 3: understand what the words are supposed to be. But it

845
00:50:11,560 --> 00:50:15,920
Speaker 3: doesn't necessarily mean that people perceive these machines as speaking,

846
00:50:16,520 --> 00:50:20,239
Speaker 3: because there's another important quality to speech that was not

847
00:50:20,320 --> 00:50:23,160
Speaker 3: really captured by these early machines, and you could argue

848
00:50:23,239 --> 00:50:27,080
Speaker 3: is still somewhat lacking in the best speech synthesis of today,

849
00:50:27,520 --> 00:50:31,640
Speaker 3: and that is the natural character of continuous speech. These

850
00:50:31,680 --> 00:50:36,520
Speaker 3: machines always produce speech that sounded stilted, unreal, alien. It

851
00:50:36,560 --> 00:50:39,400
Speaker 3: was never something that would make you feel like you

852
00:50:39,440 --> 00:50:42,799
Speaker 3: were actually being talked to, as much as sort of

853
00:50:42,880 --> 00:50:48,759
Speaker 3: receiving a weird alien code in your language. And here

854
00:50:48,800 --> 00:50:51,560
Speaker 3: I just want to read from the stories chapter quote.

855
00:50:52,200 --> 00:50:56,000
Speaker 3: As a result, synthesis often presents itself as an oral

856
00:50:56,320 --> 00:50:59,880
Speaker 3: caricature that can be perceived as an unnatural in some

857
00:51:00,040 --> 00:51:03,920
Speaker 3: times amusing rendition of a desired utterance or speech sound.

858
00:51:04,280 --> 00:51:08,560
Speaker 3: It is particularly unique to phonetics and speech science that

859
00:51:08,600 --> 00:51:13,080
Speaker 3: the models used as tools to understand the scientific aspects

860
00:51:13,120 --> 00:51:17,040
Speaker 3: of a complex system produce a signal intended to be

861
00:51:17,160 --> 00:51:20,080
Speaker 3: heard as if it were a human. As such, the

862
00:51:20,160 --> 00:51:23,800
Speaker 3: quality of a speech synthesis can be rather harshly judged

863
00:51:23,920 --> 00:51:27,040
Speaker 3: because the model on which it is based has not

864
00:51:27,080 --> 00:51:31,000
Speaker 3: accounted for the myriad of subtle variations and details that

865
00:51:31,120 --> 00:51:35,880
Speaker 3: combine in natural human speech. So to paraphrase, speech is

866
00:51:35,960 --> 00:51:39,520
Speaker 3: so much more than just the words, And even if

867
00:51:39,600 --> 00:51:42,840
Speaker 3: you can get the words right, there's still something that

868
00:51:43,120 --> 00:51:45,719
Speaker 3: is that is lacking and is going to take a

869
00:51:45,840 --> 00:51:47,720
Speaker 3: lot of work to try to capture.

870
00:51:48,080 --> 00:51:51,279
Speaker 1: Yeah, this is fascinating to think about, and especially given

871
00:51:51,880 --> 00:51:55,080
Speaker 1: what you mentioned earlier about it's the importance of speech

872
00:51:55,280 --> 00:51:59,520
Speaker 1: the synthesizer technology to aid people who cannot speak or

873
00:51:59,560 --> 00:52:03,280
Speaker 1: have lost ability to speak. You know, I gave probably

874
00:52:03,560 --> 00:52:05,400
Speaker 1: one of the most famous, if not the most famous

875
00:52:05,440 --> 00:52:09,359
Speaker 1: examples of this is, of course, the speech synthesizer used

876
00:52:09,400 --> 00:52:15,080
Speaker 1: by theoretical Stephen Hawking. Like one of the interesting things

877
00:52:15,120 --> 00:52:17,520
Speaker 1: about his story with it, as I remember, is that

878
00:52:18,040 --> 00:52:20,359
Speaker 1: just me mentioning it, you can probably sort of hear

879
00:52:20,400 --> 00:52:23,880
Speaker 1: the voice the synthesized voice of Stephen Hawking in your head.

880
00:52:24,560 --> 00:52:27,120
Speaker 1: And I know that at some point like that was

881
00:52:27,239 --> 00:52:30,239
Speaker 1: you know, an early system he got there, and later

882
00:52:30,320 --> 00:52:32,480
Speaker 1: on in life he had he could have switched the

883
00:52:32,560 --> 00:52:35,640
Speaker 1: voice up, he could have changed the voice and and

884
00:52:35,800 --> 00:52:38,520
Speaker 1: I'm assuming could have maybe improved upon it, but by

885
00:52:38,560 --> 00:52:41,520
Speaker 1: that point he felt that this was his voice. You know,

886
00:52:41,560 --> 00:52:43,719
Speaker 1: you can't switch it up. You know, this is this

887
00:52:43,800 --> 00:52:46,600
Speaker 1: is how I speak, and this is how I hear myself.

888
00:52:47,360 --> 00:52:49,839
Speaker 1: So I always found that that interesting, and especially when

889
00:52:49,920 --> 00:52:51,840
Speaker 1: and then you can compare that to some other cases,

890
00:52:51,880 --> 00:52:54,799
Speaker 1: like you know, film credit Roger Ebert late in life,

891
00:52:54,840 --> 00:52:56,960
Speaker 1: you know, you could no longer speak, but had I

892
00:52:56,960 --> 00:52:59,799
Speaker 1: think they had a more robust system put together based

893
00:52:59,840 --> 00:53:03,239
Speaker 1: on samples of you know, the great catalog of his

894
00:53:03,320 --> 00:53:07,680
Speaker 1: own recorded speeches and reviews and so forth that they

895
00:53:07,680 --> 00:53:10,160
Speaker 1: could draw upon. And then looking into the future, you

896
00:53:10,160 --> 00:53:14,160
Speaker 1: have situations like James Earl Jones's Darth Vader voice, that

897
00:53:14,280 --> 00:53:21,200
Speaker 1: being you know, sort of archived and prepared for so

898
00:53:21,239 --> 00:53:23,839
Speaker 1: that in the future you can you can basically have

899
00:53:24,200 --> 00:53:28,120
Speaker 1: like a machine synthesized version of that voice that will

900
00:53:28,160 --> 00:53:31,600
Speaker 1: stand in as a sort of one to one replication

901
00:53:31,760 --> 00:53:34,239
Speaker 1: of what James Earl Jones did in life with the

902
00:53:34,360 --> 00:53:35,319
Speaker 1: voice acting.

903
00:53:36,280 --> 00:53:38,680
Speaker 3: Or at least so the proponents of the technology would say,

904
00:53:38,680 --> 00:53:40,440
Speaker 3: I'm sure there would be critics who would say, it's

905
00:53:40,480 --> 00:53:41,920
Speaker 3: never going to be a one to one.

906
00:53:42,440 --> 00:53:44,600
Speaker 1: Right, right, And then of course there's also the argument,

907
00:53:44,800 --> 00:53:48,920
Speaker 1: specifically with only with Darth Vader. Here am I discussing this,

908
00:53:49,040 --> 00:53:52,239
Speaker 1: but obviously the case can be made that like, well,

909
00:53:52,280 --> 00:53:57,160
Speaker 1: we shouldn't reproduce, you know, deceased actors' voices to continue

910
00:53:57,160 --> 00:54:01,600
Speaker 1: a fictional role. We should employ new living actors and

911
00:54:01,719 --> 00:54:04,799
Speaker 1: existing living voice actors who can do the voice. I

912
00:54:04,800 --> 00:54:07,400
Speaker 1: think with Darth Vader in particular, you could make a

913
00:54:07,440 --> 00:54:09,600
Speaker 1: strong case for that because there are other voice actors

914
00:54:09,600 --> 00:54:13,400
Speaker 1: who do officially voice act that character and do a

915
00:54:13,440 --> 00:54:17,319
Speaker 1: great job with it. What does it mean if that

916
00:54:17,560 --> 00:54:22,120
Speaker 1: individual's job is potentially taken by this sort of machine

917
00:54:22,480 --> 00:54:26,719
Speaker 1: likeness of that voice that is authorized based on the

918
00:54:26,840 --> 00:54:31,040
Speaker 1: voice of a you know, of a retired or in

919
00:54:31,080 --> 00:54:33,040
Speaker 1: some cases you know, deceased individual.

920
00:54:33,320 --> 00:54:35,600
Speaker 3: Well, we're going a little off topic now, but I

921
00:54:35,640 --> 00:54:37,919
Speaker 3: will say that I stand by what I've said before,

922
00:54:37,960 --> 00:54:40,480
Speaker 3: which is I'm firmly in the camp that I prefer

923
00:54:40,760 --> 00:54:45,040
Speaker 3: recasting with a different actor, as opposed to using technology

924
00:54:45,080 --> 00:54:48,000
Speaker 3: to try to synthesize the voice or appearance of an

925
00:54:48,000 --> 00:54:51,520
Speaker 3: actor who, for whatever reason cannot be present. Right, people

926
00:54:51,600 --> 00:54:54,600
Speaker 3: have been recasting the same role with different actors for decades.

927
00:54:54,640 --> 00:54:56,719
Speaker 3: That happens all the time. Like, what's the problem with it?

928
00:54:57,280 --> 00:55:00,680
Speaker 1: Yeah? I agree? I agree? But in in some cases,

929
00:55:00,719 --> 00:55:04,560
Speaker 1: is it possible that a role that's been established by

930
00:55:05,040 --> 00:55:10,719
Speaker 1: by a living actor could not be just masterfully redone

931
00:55:11,280 --> 00:55:15,560
Speaker 1: by a clunky machine with the face of a cherub,

932
00:55:15,960 --> 00:55:19,200
Speaker 1: that is, that is manipulated by a sad German man

933
00:55:19,239 --> 00:55:21,879
Speaker 1: who needs a haircut. I think there's some potential there,

934
00:55:21,960 --> 00:55:24,040
Speaker 1: Like I don't know the next James Bond.

935
00:55:24,080 --> 00:55:27,480
Speaker 3: Maybe this is the only film genre I'm interested in

936
00:55:27,520 --> 00:55:32,680
Speaker 3: from now on. Yeah, high tension espionage movies starring the euphonia.

937
00:55:34,480 --> 00:55:37,480
Speaker 1: So there you have it. The machine speaks. Obviously, we'd

938
00:55:37,520 --> 00:55:38,840
Speaker 1: love to hear from everyone out there if you have

939
00:55:38,880 --> 00:55:40,880
Speaker 1: thoughts on all of this, and certainly anyone out there

940
00:55:40,920 --> 00:55:44,879
Speaker 1: who has you know, direct experience with speech synthesizer technology

941
00:55:45,680 --> 00:55:48,160
Speaker 1: for one use or another. Right in, we would love

942
00:55:48,200 --> 00:55:48,880
Speaker 1: to hear from you.

943
00:55:49,520 --> 00:55:54,720
Speaker 3: Just a reminder, I just the speech synthesis or speech

944
00:55:54,800 --> 00:55:58,600
Speaker 3: synthesizer is one of the hardest pairs of words to enunciate,

945
00:55:58,640 --> 00:56:00,160
Speaker 3: and I've had to say it so many times in

946
00:56:00,160 --> 00:56:04,040
Speaker 3: this episode. I just want to be recognized, especially for

947
00:56:04,080 --> 00:56:05,319
Speaker 3: the times I probably did it wrong.

948
00:56:06,160 --> 00:56:09,040
Speaker 1: Yes, well it's easy for the babyface machines that yeah.

949
00:56:09,120 --> 00:56:11,719
Speaker 1: So at any rate. Yeah. If you want to listen

950
00:56:11,719 --> 00:56:13,320
Speaker 1: to other episodes of Stuff to Blow Your Mind, you

951
00:56:13,360 --> 00:56:14,680
Speaker 1: will find them in the Stuff to Blow Your Mind

952
00:56:14,680 --> 00:56:17,759
Speaker 1: podcast feed with our core episodes on Tuesdays and Thursdays.

953
00:56:18,120 --> 00:56:20,520
Speaker 1: Mondays we do a listener mail, Wednesdays we do a

954
00:56:20,560 --> 00:56:23,080
Speaker 1: short form artufactor monster fact, and then on Fridays we

955
00:56:23,160 --> 00:56:25,719
Speaker 1: set aside most serious concerns to just talk about a

956
00:56:25,760 --> 00:56:27,680
Speaker 1: weird film on Weird House Cinema.

957
00:56:27,840 --> 00:56:31,799
Speaker 3: Huge thanks to our excellent audio producer JJ Posway. If

958
00:56:31,840 --> 00:56:33,239
Speaker 3: you would like to get in touch with us with

959
00:56:33,280 --> 00:56:36,120
Speaker 3: feedback on this episode or any other, to suggest topic

960
00:56:36,160 --> 00:56:38,040
Speaker 3: for the future, or just to say hello, you can

961
00:56:38,120 --> 00:56:48,680
Speaker 3: email us at contact Stuff to Blow your Mind dot com.

962
00:56:48,800 --> 00:56:51,759
Speaker 2: Stuff to Blow Your Mind is production of iHeartRadio. For

963
00:56:51,840 --> 00:56:54,640
Speaker 2: more podcasts from my Heart Radio, visit the iHeartRadio app,

964
00:56:54,760 --> 00:57:12,080
Speaker 2: Apple Podcasts, or wherever you listen to your favorite shows,