1
00:00:00,000 --> 00:00:02,422
Text to Speech Emotion, Set Up for You in

2
00:00:02,462 --> 00:00:03,107
One Click.

3
00:00:04,707 --> 00:00:06,952
Text to speech emotion is what turns a

4
00:00:06,992 --> 00:00:09,413
flat reading into a voiceover people keep

5
00:00:09,453 --> 00:00:10,215
listening to.

6
00:00:10,615 --> 00:00:12,609
In most AI voice tools you get it by

7
00:00:12,649 --> 00:00:14,804
writing style prompts, rendering a take,

8
00:00:14,844 --> 00:00:17,053
listening and trying again until the tone

9
00:00:17,093 --> 00:00:17,789
sounds right.

10
00:00:18,189 --> 00:00:20,603
TTSLark does that part for you.

11
00:00:21,003 --> 00:00:23,102
Tell it what you are making, paste your

12
00:00:23,142 --> 00:00:25,346
script and it sets the tone and pacing for

13
00:00:25,386 --> 00:00:27,212
each part before any audio is made.

14
00:00:27,612 --> 00:00:30,483
You get a finished MP3 with captions in

15
00:00:30,523 --> 00:00:31,233
one pass.

16
00:00:33,833 --> 00:00:36,469
Why is text to speech emotion so much work

17
00:00:36,509 --> 00:00:37,443
in most tools?

18
00:00:39,043 --> 00:00:41,388
A voice model can only act as well as the

19
00:00:41,428 --> 00:00:43,603
direction it gets, and most tools hand

20
00:00:43,643 --> 00:00:44,551
that job to you.

21
00:00:44,951 --> 00:00:47,078
You write a style prompt, pick emotion

22
00:00:47,118 --> 00:00:49,408
settings line by line, render a sample and

23
00:00:49,448 --> 00:00:49,827
listen.

24
00:00:50,227 --> 00:00:52,169
If the second paragraph comes out too

25
00:00:52,209 --> 00:00:54,253
cheerful, you rewrite the prompt and pay

26
00:00:54,293 --> 00:00:55,259
for another render.

27
00:00:55,659 --> 00:00:57,718
A long script can take an afternoon of

28
00:00:57,758 --> 00:00:59,752
this before you have one usable take.

29
00:01:00,352 --> 00:01:02,814
That is why comparing tools by voice count

30
00:01:02,854 --> 00:01:03,843
misses the point.

31
00:01:04,243 --> 00:01:06,380
A library of thousands of voices still

32
00:01:06,420 --> 00:01:08,230
gives you a flat reading when the

33
00:01:08,270 --> 00:01:09,304
direction is wrong.

34
00:01:09,704 --> 00:01:11,775
What most creators need is the shortest

35
00:01:11,815 --> 00:01:13,990
path from a finished script to a voiceover

36
00:01:14,030 --> 00:01:15,266
they can publish today.

37
00:01:17,866 --> 00:01:20,251
How does TTSLark set text to speech

38
00:01:20,291 --> 00:01:21,436
emotion for you?

39
00:01:23,036 --> 00:01:25,440
AI Voice Over turns direction into a

40
00:01:25,480 --> 00:01:26,868
choice you make once.

41
00:01:27,268 --> 00:01:29,380
The whole process takes five steps.

42
00:01:29,980 --> 00:01:30,955
Step one.

43
00:01:31,355 --> 00:01:32,826
Say what you are making.

44
00:01:33,226 --> 00:01:35,620
Choose YouTube video, Podcast, Story,

45
00:01:35,660 --> 00:01:38,241
Audiobook, Ad, Course, Documentary, News,

46
00:01:38,281 --> 00:01:40,402
Meditation, Kids story or Trailer.

47
00:01:40,802 --> 00:01:42,761
For anything else, pick Something else

48
00:01:42,801 --> 00:01:44,859
and describe the sound in your own words,

49
00:01:44,899 --> 00:01:45,748
in any language.

50
00:01:46,348 --> 00:01:47,325
Step two.

51
00:01:47,725 --> 00:01:49,177
Add a mood if you like.

52
00:01:49,577 --> 00:01:51,977
Suspense, Sad, Joyful, Excited, Angry,

53
00:01:52,017 --> 00:01:54,540
Calm or Romantic shifts the whole reading

54
00:01:54,580 --> 00:01:55,312
in one tap.

55
00:01:55,912 --> 00:01:56,959
Step three.

56
00:01:57,359 --> 00:01:59,330
Paste your script and pick a voice.

57
00:01:59,730 --> 00:02:01,278
Play a sample first.

58
00:02:01,678 --> 00:02:03,721
Hear the difference lets you compare a

59
00:02:03,761 --> 00:02:05,583
plain reading with a directed one.

60
00:02:06,183 --> 00:02:07,166
Step four.

61
00:02:07,566 --> 00:02:08,759
Set up the reading.

62
00:02:09,159 --> 00:02:11,688
TTSLark splits your script into parts and

63
00:02:11,728 --> 00:02:14,117
gives each part its own tone and pacing.

64
00:02:14,517 --> 00:02:16,980
This step is free and no audio is made

65
00:02:17,020 --> 00:02:17,333
yet.

66
00:02:17,933 --> 00:02:18,937
Step five.

67
00:02:19,337 --> 00:02:20,548
Create audio.

68
00:02:20,948 --> 00:02:23,146
You get one MP3 for the whole script,

69
00:02:23,186 --> 00:02:25,309
plus captions for your video editor.

70
00:02:25,909 --> 00:02:28,290
Text to speech emotion here follows the

71
00:02:28,330 --> 00:02:29,178
script itself.

72
00:02:29,578 --> 00:02:32,123
A story rises and falls with the plot.

73
00:02:32,523 --> 00:02:34,902
A course stays patient and clear, like a

74
00:02:34,942 --> 00:02:35,691
good teacher.

75
00:02:36,091 --> 00:02:38,729
An ad stays bright and ends with a punch.

76
00:02:39,129 --> 00:02:40,964
A trailer builds to a peak.

77
00:02:41,364 --> 00:02:43,580
You choose the kind of project and never

78
00:02:43,620 --> 00:02:44,480
write a prompt.

79
00:02:47,080 --> 00:02:49,304
How do you make an AI voice sound more

80
00:02:49,344 --> 00:02:49,750
human?

81
00:02:51,350 --> 00:02:53,463
An AI voice sounds more human when its

82
00:02:53,503 --> 00:02:55,781
delivery changes with the meaning of each

83
00:02:55,821 --> 00:02:57,753
line, the way a person reads aloud.

84
00:02:58,153 --> 00:03:00,268
One fixed style for a whole script is

85
00:03:00,308 --> 00:03:02,519
where most robotic voiceovers come from.

86
00:03:02,919 --> 00:03:05,127
Because each part of your script gets its

87
00:03:05,167 --> 00:03:07,271
own direction, a tense moment slows down

88
00:03:07,311 --> 00:03:09,403
and a punchline lands with more energy.

89
00:03:10,003 --> 00:03:12,299
You still have the final say, with buttons

90
00:03:12,339 --> 00:03:13,425
instead of prompts.

91
00:03:14,025 --> 00:03:16,339
Calmer overall, More energy overall or

92
00:03:16,379 --> 00:03:18,934
Slower overall changes the whole script in

93
00:03:18,974 --> 00:03:19,638
one click.

94
00:03:20,238 --> 00:03:22,309
Review each part lets you make one part

95
00:03:22,349 --> 00:03:24,525
calmer, give it more energy, slow it down

96
00:03:24,565 --> 00:03:26,043
or try a different reading.

97
00:03:26,643 --> 00:03:29,004
Edit with tags places a pause, a laugh, a

98
00:03:29,044 --> 00:03:31,406
sigh or a breath exactly where you want it

99
00:03:31,446 --> 00:03:31,560
.

100
00:03:32,160 --> 00:03:34,412
These changes cost nothing until you

101
00:03:34,452 --> 00:03:35,257
create audio.

102
00:03:35,657 --> 00:03:37,665
You hear what you are paying for before

103
00:03:37,705 --> 00:03:39,412
you spend a credit, so there is no

104
00:03:39,452 --> 00:03:41,000
guessing and no wasted render.

105
00:03:43,600 --> 00:03:45,839
Do you need AI voice prompts at all?

106
00:03:47,439 --> 00:03:48,046
No.

107
00:03:48,446 --> 00:03:50,645
Prompt writing is a skill of its own, and

108
00:03:50,685 --> 00:03:52,831
the words that work change from one model

109
00:03:52,871 --> 00:03:53,495
to the next.

110
00:03:53,895 --> 00:03:56,171
With TTSLark the direction is built from

111
00:03:56,211 --> 00:03:58,322
your choices: what you are making, the

112
00:03:58,362 --> 00:03:59,409
mood and the voice.

113
00:03:59,809 --> 00:04:01,992
That is the fastest way we know to get

114
00:04:02,032 --> 00:04:04,326
text to speech emotion right on the first

115
00:04:04,366 --> 00:04:04,811
render.

116
00:04:05,411 --> 00:04:07,501
If you already know exactly how a script

117
00:04:07,541 --> 00:04:09,578
should sound, choose Something else and

118
00:04:09,618 --> 00:04:11,552
describe it in a sentence, such as "A

119
00:04:11,592 --> 00:04:13,358
bedtime story for a five year old.

120
00:04:13,758 --> 00:04:15,931
Gentle, and a little sleepier toward the

121
00:04:15,971 --> 00:04:16,234
end."

122
00:04:16,634 --> 00:04:19,083
That one sentence does the job of a page

123
00:04:19,123 --> 00:04:20,012
of style notes.

124
00:04:20,412 --> 00:04:23,143
TTSLark turns it into direction for every

125
00:04:23,183 --> 00:04:24,472
part of the script.

126
00:04:27,072 --> 00:04:29,344
What makes natural sounding text to speech

127
00:04:29,384 --> 00:04:30,567
with standard voices?

128
00:04:32,167 --> 00:04:34,150
Not every video needs acting.

129
00:04:34,550 --> 00:04:36,725
An explainer or a tutorial calls for a

130
00:04:36,765 --> 00:04:39,051
clear narrator at a steady pace, and that

131
00:04:39,091 --> 00:04:41,084
is what the standard voices do best.

132
00:04:41,484 --> 00:04:43,871
The free text to speech generator has

133
00:04:43,911 --> 00:04:45,986
more than 50 standard voices in 9

134
00:04:46,026 --> 00:04:46,649
languages.

135
00:04:47,049 --> 00:04:49,146
Paste the script, pick a voice, set the

136
00:04:49,186 --> 00:04:51,232
speed anywhere from half speed to twice

137
00:04:51,272 --> 00:04:52,732
the normal speed and render.

138
00:04:53,132 --> 00:04:55,634
A long script comes back as one

139
00:04:55,674 --> 00:04:56,906
continuous MP3.

140
00:04:57,506 --> 00:05:00,286
Every render includes SRT and VTT files

141
00:05:00,326 --> 00:05:03,145
built from the same render as the audio.

142
00:05:03,545 --> 00:05:05,676
English standard voices get word level

143
00:05:05,716 --> 00:05:07,846
timing, so the captions match the voice

144
00:05:07,886 --> 00:05:09,622
from the first line to the last.

145
00:05:10,022 --> 00:05:12,700
For a YouTube video, drop the MP3 on your

146
00:05:12,740 --> 00:05:15,102
timeline, import the SRT file and the

147
00:05:15,142 --> 00:05:16,279
captions are done.

148
00:05:16,679 --> 00:05:18,980
No transcription pass and no fixing

149
00:05:19,020 --> 00:05:20,032
misheard names.

150
00:05:20,632 --> 00:05:22,930
Pacing and pronunciation stay in the

151
00:05:22,970 --> 00:05:23,412
script.

152
00:05:23,812 --> 00:05:25,876
Add a two second pause for a beat, or

153
00:05:25,916 --> 00:05:28,142
select a word and use Pronounce as to set

154
00:05:28,182 --> 00:05:29,099
how it is spoken.

155
00:05:29,499 --> 00:05:31,841
This guide is read by Miles, a standard

156
00:05:31,881 --> 00:05:33,933
voice we noted for narrated product

157
00:05:33,973 --> 00:05:34,902
demonstrations.

158
00:05:37,502 --> 00:05:39,541
How does Voice Mix give your channel a

159
00:05:39,581 --> 00:05:40,540
voice of its own?

160
00:05:42,140 --> 00:05:44,675
Voice Mix blends any two standard voices

161
00:05:44,715 --> 00:05:45,280
into one.

162
00:05:45,680 --> 00:05:47,474
Choose the voice that sets the

163
00:05:47,514 --> 00:05:49,823
pronunciation, then set how much of each

164
00:05:49,863 --> 00:05:52,000
you hear, from 10 to 90 percent in 10

165
00:05:52,040 --> 00:05:52,842
percent steps.

166
00:05:53,242 --> 00:05:55,431
Nine settings for every pair add up to

167
00:05:55,471 --> 00:05:57,660
thousands of distinct voices, and every

168
00:05:57,700 --> 00:05:59,945
blend plays at any speed from half speed

169
00:05:59,985 --> 00:06:01,490
to twice the normal speed.

170
00:06:02,090 --> 00:06:04,425
Preview any blend for free, save it under

171
00:06:04,465 --> 00:06:06,783
your own name and use it in every script.

172
00:06:07,183 --> 00:06:09,335
If you want a shortcut, start from one of

173
00:06:09,375 --> 00:06:11,220
the ready-made blends and adjust the

174
00:06:11,260 --> 00:06:11,973
weight by ear.

175
00:06:12,373 --> 00:06:14,626
No recording is needed, and your channel

176
00:06:14,666 --> 00:06:16,701
ends up with a narrator that no other

177
00:06:16,741 --> 00:06:17,614
channel shares.

178
00:06:18,214 --> 00:06:20,589
This is the free side of text to speech

179
00:06:20,629 --> 00:06:21,112
emotion.

180
00:06:21,512 --> 00:06:23,465
A blend gives you a voice with its own

181
00:06:23,505 --> 00:06:25,458
character, and that character stays the

182
00:06:25,498 --> 00:06:27,202
same from your first video to your

183
00:06:27,242 --> 00:06:27,790
hundredth.

184
00:06:30,390 --> 00:06:32,272
Which path fits your project?

185
00:06:33,872 --> 00:06:36,436
For explainers, tutorials and long videos,

186
00:06:36,476 --> 00:06:38,858
use the standard voices and get a clear

187
00:06:38,898 --> 00:06:40,412
MP3 with synced captions.

188
00:06:40,812 --> 00:06:43,108
For a channel voice nobody else has, use

189
00:06:43,148 --> 00:06:45,038
Voice Mix and save your own blend.

190
00:06:45,438 --> 00:06:47,892
For stories, ads, trailers and audiobooks

191
00:06:47,932 --> 00:06:50,444
, use AI Voice Over and get acted delivery

192
00:06:50,484 --> 00:06:51,434
set up for you.

193
00:06:52,034 --> 00:06:54,464
All three paths end the same way: an MP3

194
00:06:54,504 --> 00:06:56,813
and caption files you can put straight

195
00:06:56,853 --> 00:06:57,877
into your editor.

196
00:06:58,277 --> 00:07:00,934
Commercial use is included on every plan,

197
00:07:00,974 --> 00:07:03,358
with no watermark and no attribution.

198
00:07:05,958 --> 00:07:07,644
Frequently asked questions.

199
00:07:10,244 --> 00:07:12,327
Can text to speech show emotion?

200
00:07:12,927 --> 00:07:13,612
Yes.

201
00:07:14,012 --> 00:07:16,647
Premium voices on AI Voice Over act out

202
00:07:16,687 --> 00:07:17,405
the script.

203
00:07:17,805 --> 00:07:19,613
They change tone and pace with the

204
00:07:19,653 --> 00:07:21,718
meaning of each line and can laugh, sigh

205
00:07:21,758 --> 00:07:23,350
or pause where you place a tag.

206
00:07:23,750 --> 00:07:26,267
TTSLark sets that delivery for each part

207
00:07:26,307 --> 00:07:28,134
based on what you are making.

208
00:07:29,734 --> 00:07:32,249
How do I add emotion to text to speech?

209
00:07:32,849 --> 00:07:35,375
Open AI Voice Over and choose what you are

210
00:07:35,415 --> 00:07:35,833
making.

211
00:07:36,233 --> 00:07:38,421
Paste your script and select Set up the

212
00:07:38,461 --> 00:07:38,895
reading.

213
00:07:39,295 --> 00:07:41,205
Each part gets its own tone.

214
00:07:41,605 --> 00:07:43,711
You can also add a mood for the whole

215
00:07:43,751 --> 00:07:46,077
script or insert tags such as a laugh or a

216
00:07:46,117 --> 00:07:46,942
pause by hand.

217
00:07:48,542 --> 00:07:50,779
Does setting up the reading cost credits?

218
00:07:51,379 --> 00:07:51,977
No.

219
00:07:52,377 --> 00:07:54,607
Setting up the reading and adjusting it

220
00:07:54,647 --> 00:07:55,146
are free.

221
00:07:55,546 --> 00:07:57,783
Credits are used only when you create

222
00:07:57,823 --> 00:07:58,173
audio.

223
00:07:58,573 --> 00:08:01,778
Premium voices use 1 credit per 400 bytes

224
00:08:01,818 --> 00:08:02,422
of text.

225
00:08:02,822 --> 00:08:05,299
Standard voices use 1 credit per 10,000

226
00:08:05,339 --> 00:08:07,693
bytes, which makes up to 10 minutes of

227
00:08:07,733 --> 00:08:08,162
audio.

228
00:08:09,762 --> 00:08:11,263
Which plan do I need?

229
00:08:11,863 --> 00:08:14,235
Standard voices, Voice Mix and captions

230
00:08:14,275 --> 00:08:15,542
come with every plan.

231
00:08:15,942 --> 00:08:18,460
The free plan gives you five credits that

232
00:08:18,500 --> 00:08:19,273
never expire.

233
00:08:19,673 --> 00:08:22,090
AI Voice Over is part of Starter and

234
00:08:22,130 --> 00:08:22,712
Creator.

235
00:08:25,312 --> 00:08:27,042
Try it with your next script.

236
00:08:28,642 --> 00:08:30,820
Paste a script into AI Voice Over, choose

237
00:08:30,860 --> 00:08:33,038
what you are making and set up the reading

238
00:08:33,078 --> 00:08:33,131
.

239
00:08:33,531 --> 00:08:35,424
You hear the text to speech emotion

240
00:08:35,464 --> 00:08:36,823
before you spend a credit.

241
00:08:37,223 --> 00:08:39,466
For a plain narrator with captions, the

242
00:08:39,506 --> 00:08:41,860
free text to speech generator is one paste

243
00:08:41,900 --> 00:08:42,234
away.
