YouTube2Text

How I Actually Make AI Voice Sound Real — Transcript

by Isaac · 3,154 words · 230 segments · language en · Watch on YouTube

Full transcript

  1. 0:10So, in this vid- no, wait. So, in this video,  we’re going to… break down the… Ugh, that
  2. 0:17sounds terrible… Still recording?
  3. 0:22I can’t get it right. I just  don’t sound… like a youtuber.
  4. 0:28Good.
  5. 0:29Good? What’s good about that? Have you heard an advice,
  6. 0:32“if you’re homeless, just buy a house”. What the f? That doesn’t make any sense.
  7. 0:37I mean if you don’t sound like a youtuber,  then just use something that does.
  8. 0:44And just like that, I went down the  rabbit hole of… AI voice generators.
  9. 1:08So… this is how I actually  make an AI voice sound… real…
  10. 1:17Okay, look. Before we get into the technical  stuff, if you wanna make a realistic voice with
  11. 1:22AI, you need to understand what the hell makes  human voice realistic in the first place. Really,
  12. 1:27like why does your brain think you're listening  to a real person right now instead of a computer
  13. 1:32pretending to be a ginger dude with a fancy suit?  Well, if you break it down, there are 4 elements.
  14. 1:37If I keep the same tone throughout the entire  video with no ups and downs in my voice,
  15. 1:42that would make it sound robotic. I  talk like that in real life… Oh, man,
  16. 1:46this robot can talk more human than me?  Anyways! If all of a sudden I change my
  17. 1:50tone from monotonous to super excited,  that would make it sound more human.
  18. 1:57You probably heard that cutting all the pauses  in your video boosts retention. Even though
  19. 2:02that’s kinda right, it’s not worth making  yourself look like a lifeless robot that’s
  20. 2:06been controlled by the algorithms. Instead,  pause on some important points… to make your
  21. 2:11message… more impactful… And make your  AI voice sound more natural, of course.
  22. 2:19Just like making pauses, emphasizing important  words in your script… no… EMPHASIZING important
  23. 2:24words in your script, is also a powerful way  to make your AI voice sound more realistic.
  24. 2:32And lastly, of course, if you use  ChatGPT to write your entire script,
  25. 2:36no matter how realistic your AI voice is, at some  point, it will start to sound robotic. And yeah,
  26. 2:42if you write your scripts with AI, and  think that nobody knows… trust me… we know…
  27. 2:48Alright, so with these 4 elements, you  can cook a 96.2% pure ai voice. These
  28. 2:53are the main things you need to keep in  mind while creating your voiceovers. And
  29. 2:57we’ll use each of this in a minute  while creating the actual voiceover.
  30. 3:00But before that, we need to…
  31. 3:03Choose a software.
  32. 3:04Look, there are so many AI voice generators out  there right now that I just gave up counting. I
  33. 3:09tested almost every single one for this  video, and you know what? Most of them
  34. 3:13are… fine. They do a decent job at replicating a  human voice. But the one I always come back to,
  35. 3:19is called, ElevenLabs. So that’s gonna  be the software we’ll use for this video.
  36. 3:23And no, they didn’t pay me to say  this. I wish they did, but they didn’t.
  37. 3:27I’ve been using it since I started this channel,
  38. 3:29and you’re seeing… or hearing how  realistic it sounds. Be honest. You
  39. 3:34sometimes forget it’s not a human voice,  right? You… you don’t? Okay, nevermind…
  40. 3:39There’s a link in the description if you wanna
  41. 3:41check it out and follow along with your  own script. Just create and account,
  42. 3:44and try the techniques we’re gonna use  by yourself. Okay, fine. Let’s do this.
  43. 3:49So you know the theory. Now,  what do you need? A voice!
  44. 3:53This is an important step. Because once you  pick a voice, you're basically married to
  45. 3:57it. All your future videos. That voice. So  don't just click the first one you see and
  46. 4:01call it a day. You’re gonna need something  solid. Take your time. Pick a nice one.
  47. 4:05Now there are a few ways to create a voice.
  48. 4:08First, if you wanna use your own  voice, you can go with this route.
  49. 4:12You may think “why would I clone my own  voice when I can just… use my throat”,
  50. 4:17but it’s gonna be a really important step  for the end of the video. Just let me cook.
  51. 4:21Click the ‘voices’ tab. ‘clone a voice’, then  ‘instant voice clone’, and upload a sample.
  52. 4:29If you got time, you can create a  professional voice clone as well.
  53. 4:32But keep in mind that it requires at  least 30 minutes of voice sample. Once
  54. 4:36you generated your voice clone, “you can  use it wherever you want. that’s crazy”.
  55. 4:40Wait, actually… yeah, that’s completely illegal.
  56. 4:44You can’s clone someone else’s voice  without their permission. It’s like,
  57. 4:47against every terms of service ever written.  So you better use it with your own voice only.
  58. 4:52And, uh, Finzar, if you’re watching  this… please don’t sue me, dude.
  59. 4:56get the hell out of here!
  60. 4:59seems like no ai can say it like the original
  61. 5:04Now if you don’t like your own voice,
  62. 5:05you can choose one of the voices  from the huge voice library here.
  63. 5:20Anyways, so, the most asked  question in this channel,
  64. 5:22after “how you edit your videos”, is “which  voice are you using in ElevenLabs”. Well,
  65. 5:27the voice you’re hearing right now, isn’t  from the voice library or a cloned voice…
  66. 5:32It’s a custom voice! Most tutorials don’t mention  this, but you can actually create an entirely new
  67. 5:39unique voice in ElevenLabs. It’s usually as  realistic as the ones from the voice library,
  68. 5:44while being not available to everyone with  one click. So it’s a great choice for creating
  69. 5:48a brand “that doesn’t use this voice”. No  one else can use it for their channel. Uhm,
  70. 5:52well, if they don’t clone your voice with the  cloning feature, but you get my point, right?
  71. 5:57In the ‘voices’ tab, click ‘create a voice’,  ‘voice design’. When I created my voice a few
  72. 6:02years ago, this feature was just a single button  that you can click, and hope for a good voice.
  73. 6:07But now, you can describe the voice in your mind  however you want. Like “a young man in his 20s,
  74. 6:12american accent, speaking with quirky, but  charismatic style. speaks with a lot of emotions
  75. 6:15and variances in his speech, changes the tone  and speed continuously throughout the speech”. I
  76. 6:18don’t know, something like that. And let’s reduce  the guidance a bit to see what it comes up with.
  77. 6:33Damn, can I change my voice to that? No? Okay…
  78. 6:37Now, if you found a voice that  works for you, we can move on to…
  79. 6:41Generating the actual voiceover.
  80. 6:44At this point, you have a nice voice, and you know  what realistic voice sounds like. So you can just
  81. 6:48copy your script, paste it to Elevenlabs, and hit  ‘generate’ wait, hold on for a second. Of course,
  82. 6:54you can do it. You may even get a pretty good  result from it as well. But if you do that,
  83. 6:58after listening to it for a few minutes,  something will feel off. And the viewers
  84. 7:03gonna feel it as well. I don’t know what it  is, but it’s there in all AI voices. It’s like
  85. 7:07they follow a pattern that just… gives it away…  And once you notice it, you can’t unnotice it.
  86. 7:13So let me show you how I generate my voiceovers.  It takes much more time, but it’s the only way I
  87. 7:17found to get rid of that uncanny feeling. And  it’s probably the secret sauce that makes you
  88. 7:22wonder how this ginger dude sounds so realistic.  Alright, enough yapping, let’s get started.
  89. 7:27First off, put the script down slowly.
  90. 7:29We have to choose a voice model first. V3 is  their latest model. It can do crazy stuff like
  91. 7:34making the voice giggle, whisper, or be excited.  It sounds great, right? But when I tried it with
  92. 7:42my own voice… “it sounded like this”. Who the hell  is that?! That’s like… my evil twin or something.
  93. 7:48I even tried it with other voices, and  every time, it slightly changes the voice
  94. 7:53itself. Except for a few voices that are designed  for this model. So you can give them a try if you
  95. 7:55want. It’s still in the alpha stage, so hopefully  it will be more stable in the future versions.
  96. 7:59But it’s not really usable right now. So I  use the V2 model instead for all my videos.
  97. 8:04Unlike V3, this model has a few more sliders.  Increase the speed, reduce the stability,
  98. 8:09and… turn of that annoying music! I’m  trying to focus here, dude. Thank you.
  99. 8:15Drag the similarity to something like  70%, and increase the style exaggeration.
  100. 8:21Now select just a few sentences from your  script, and paste it to Elevenlabs. What
  101. 8:26did we needed for a realistic voice? Changing  tone and speed, pauses, emphasis, and writing
  102. 8:31like a human. I assume you already nailed the  last part. So let’s focus on the other three.
  103. 8:36Using 3 dots makes the voice sound disappointed,  or confused. To make it more excited,
  104. 8:41use exclamation marks. It always works.  If it doesn’t, use more exclamations!
  105. 8:46We’ll play with the pauses in the editing,
  106. 8:48but you can try 3 dots or commas  to make them a bit more natural.
  107. 8:52And for emphasizing important  words, write them in all caps.
  108. 8:55Another example, this is how it  sounded like before making the changes.
  109. 8:59“Say my name. You’re Isaac. You’re goddamn right”
  110. 9:05Pretty normal. And this is how it  sounds like after making the changes.
  111. 9:11“Say my NAME! You’re ISAAC…  You’re GODDAMN RIGHT!!”
  112. 9:16Sounds much better, but if you’re not  happy with how the result turned out,
  113. 9:19you can generate it again a few more times.
  114. 9:22“you’re goddamn right. you’re  goddamn right. you’re goddamn right”
  115. 9:26But why are we generating a few sentences at a  time? Couldn’t we do the entire script at once?
  116. 9:30Well, we could, but we shouldn’t. One reason  is, if it generates a bad take, we can easily
  117. 9:35regenerate it without wasting too much credits.  But the second reason? This is the secret sauce.
  118. 9:41If you realized, each time you hit generate,  “the tone changes slightly. the tone changes
  119. 9:45slightly. the tone changes slightly” even  though we input the same text. So, we’re
  120. 9:50taking advantage of that. For each generation,  you get 2 extra regenerations for free. Use
  121. 9:54those free generations, and download them as  well. You’re gonna see why in the editing part.
  122. 9:58With every batch we generate, we get a slightly  different tone and speed. That’s just like how
  123. 10:03normal humans sound like. So when combined  together in the edit, it makes the final
  124. 10:07result sound more natural and fixes that  uncanny feeling we talked about earlier.
  125. 10:11After knowing these, it just takes a few  hundred more generations, and you get this. Yes,
  126. 10:16it takes ages, but it’s like cooking.  Small changes make a huge difference.
  127. 10:20Now put them all in a single folder, select them  all, and rename. That’s much more organized.
  128. 10:25Now we can move on to the next step.
  129. 10:29---
  130. 10:30I’m gonna show this in Premiere Pro, but you can  do it in any editing software that has a timeline.
  131. 10:35Import the voiceovers you generated to  Premiere Pro, and start choosing the
  132. 10:38best takes. Don’t worry about timing  or anything else at this point. Just
  133. 10:42line the voiceover based on your script. If you  downloaded a few versions of the same sentence,
  134. 10:46listen to all, and choose the  one that fits the context better.
  135. 10:50But the real magic, happens when you  chop some parts from each generation,
  136. 10:53and combine them together. For this sentence, I  generated 3 versions. Let’s listen to each one.
  137. 10:58But the real magic, happens when you  chop some parts from each generation,
  138. 11:02and combine them together.
  139. 11:03But the real magic, happens when you  chop some parts from each generation,
  140. 11:07and combine them together.
  141. 11:09But the real magic, happens when you  chop some parts from each generation,
  142. 11:12and combine them together.
  143. 11:14Now they sound good on their own,  but to make it more human-like,
  144. 11:16I took the best sections from each  version, and got the final result.
  145. 11:20But the real magic, happens when you  chop some parts from each generation,
  146. 11:23and combine them together.
  147. 11:25Once you got a single long straight line,  start cutting the pauses. Use the second
  148. 11:29line to start the next sentence right after  the last one ends. But don’t forget to add
  149. 11:33pauses and intentional spaces for  the important parts of your video.
  150. 11:37Now, even though Elevenlabs had a speed slider  in its editor, you may realize just now that
  151. 11:41your voiceover is too slow, or too fast. Don’t  worry, it happens… I mean I’m not the only one,
  152. 11:47right? Premiere Pro allows you to change the  speed, but it turns you into a chipmunk if you
  153. 11:52do that. Oh, there’s an option to ‘keep  the pitch’. How’s that sound? Oh great,
  154. 11:57it distorts the voice now. Is it really  that hard to make a speed changer, Adobe?
  155. 12:01Fine, I’ll do that myself…
  156. 12:12Great, the voiceover is almost ready. But we can  make it sound more professional. Here’s something
  157. 12:17new I learned for this video. Take the parametric  equalizer effect, add a high-pass filter, increase
  158. 12:22the h value, find echo-y frequencies, drag them  down, take this slider, find a point where…
  159. 12:26Or, just drag and drop the preset I made! I  also added a few fun voice effects and other
  160. 12:31useful stuff as well for you to try out. So feel  free to download the link in the description.
  161. 12:35Well, I think we’re done here. But wait a minute…
  162. 12:39What does youtube say? Does it allow  using AI voices? Even more important,
  163. 12:45does it allow monetizing videos with AI voices?
  164. 12:48Yes.
  165. 12:49What, you expected a long list of  explanations? Haha. I mean as long
  166. 12:54as you’re making an original content, not funny  video compilations, yeah, you should be fine.
  167. 12:59But should you be fine? I mean… should we all
  168. 13:01just start using AI voices now? What  about, I don’t know… authenticity?
  169. 13:08Wow… that’s… better than I could ever do… See? Told you it would work. Sounds great, man!
  170. 13:16Yeah, but… it’s just… not me. What’s wrong, man? You’re becoming
  171. 13:22a youtuber. Wasn’t that what you wanted? I don’t know, it feels like cheating. It
  172. 13:27feels more like I’m becoming a fraud, dude. A fraud? You wanna know what’s fraud?
  173. 13:36That… Silence… Having something worth saying, and  keeping it locked up because you’re too scared of
  174. 13:43how it sounds. THAT’S fraud. Who says it should  sound like you? As long as it’s your own thoughts,
  175. 13:50who cares how it sounds, huh? Come  on, we don’t have time for this.
  176. 13:55He was right. I was so obsessed with how I sound,
  177. 13:59I was forgetting the most important  part. Actually saying something valuable…
  178. 14:05Yeah I mean I could end this  video here, and call it a day,
  179. 14:08but that’s not all. I kept  the best one for the last…
  180. 14:14I get it now. It’s a great  solution if you don’t have a mic,
  181. 14:17or if english isn’t your first language. I mean at least that’s progress,
  182. 14:22but you’re thinking too small, dude. You’re  still thinking about replacement. Don’t you
  183. 14:27see? Everything you learned, this isn’t just  about finding a voice. There’s a bigger picture.
  184. 14:32Man, talk in english, I don’t understand.
  185. 14:35Okay, name a big youtuber. I don’t know, does MrBeast count?
  186. 14:40Too cliche, but okay. When did  his videos start to get viral?
  187. 14:44I mean, he had over 10M subs in 2018. Yeah, but I mean, super viral. Look here. 2023.
  188. 14:52Yeah that’s quite a jump. Now, when did he start dubbing his videos?
  189. 14:582023. Exactly! Ruhi Cenet.
  190. 15:01Mark Rober. Joe Hattab. Anyone who started dubbing  his videos to other languages, suddenly doubled,
  191. 15:06tripled their views and subscribers. Does it really make that much difference?
  192. 15:11Think about it. There are 2.5 billion active  youtube users in the world. Around half a billion
  193. 15:16of them speaks english. That’s less than 20%. If  we just add up the next 5 most known languages
  194. 15:22to this, it makes up almost ha- Half of the entire youtube!
  195. 15:25Do you see the potential? Sheet!
  196. 15:35Well, that was the twist. AI voiceover isn’t just  for narrating your scripts for you. It’s real
  197. 15:41superpower is dubbing your videos. You can dub  your videos, and get a massive audience. To try
  198. 15:45it myself, I added just 2 languages to my last  video, and the results were pretty clear… Oh,
  199. 15:51it’s actually blurry, heheh. Don’t worry, I’ll  show it after showing how to dub your videos.
  200. 15:55As an example, let’s take my last video. I  already dubbed it into Spanish and Hindi,
  201. 16:00so let’s do Portuguese now.
  202. 16:04Before we touch any AI, we need to separate  the ingredients. After finishing your video,
  203. 16:08go to your editing timeline. Mute  everything except the voiceover and
  204. 16:11export as an MP3. Next, do the  opposite. Mute the voiceover,
  205. 16:15and export an MP3 with only the music and  sound effects. You’ll see why in a second.
  206. 16:21Now go to ElevenLabs and click the ‘Dubbing’  tab. Create a new dub. Select your original
  207. 16:25language, the target language, and upload that  voiceover we just exported. Hit generate, and
  208. 16:30go make yourself a coffee. It takes a few minutes.
  209. 16:34Once it’s done, download the file and drag  it back into Premiere Pro. Here’s what you
  210. 16:39need to build here. The top, is your  original native voice. We’re gonna use
  211. 16:42this as a guide for timing. Middle,  is the new AI Portuguese voice. And
  212. 16:47the bottom line, is for music and  sound effects we exported earlier.
  213. 16:50Line them up so the timing matches the visuals.  Once it sounds good, mute the original English
  214. 16:55track, slap my preset on the Portuguese track to  make it sound crispy, and export just the audio.
  215. 17:01Now the only thing left, is to add it to your  video. In YouTube Studio, click ‘languages’,
  216. 17:07and choose the video you just dubbed.  Make sure youtube didn’t automatically
  217. 17:11dubbed to that language with their google  translate voice. If so, just delete it from
  218. 17:15the menu here. Then click here, select the  dubbed language, and upload your audio file.
  219. 17:20Also, translate your video title and description  with ChatGPT to reach the 99.1% purity.
  220. 17:26And, you basically unlocked a new country!  Keep repeating it for other languages as well,
  221. 17:30and you tripled your potential audience size.
  222. 17:33Now as promised, here’s what it  looked like in the analytics tab…
  223. 17:37So… yeah. It doesn’t look that impressive, does  it? I mean, where’s the millions of views? Well,
  224. 17:43hold on. If you look at the views  by audio track instead… THIS,
  225. 17:46is how it looks like. Here’s Youtube’s  auto-dubbed Hindi track. And THIS,
  226. 17:51is where I added the Elevenlabs  version. That’s what I’m talking about.
  227. 17:54And this is the difference in spanish dub.
  228. 17:58Wow… that’s… thousands of people…  who watched my video! People who
  229. 18:04would have seen none of it if I had  listened to that voice in my head.
  230. 18:09Turns out, Steven was right…

About this transcript

This page contains the full transcript of How I Actually Make AI Voice Sound Real by Isaac, generated from the public captions YouTube serves with the video. The transcript has 3,154 words across 230 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to choose a caption track available on the video.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, caption language selection, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.