Speaker 1: We're in CapCut, and I've brought in a video. I've cut it down just about 10 seconds or so just for our experiment here. And this has my narration, and what we're going to do is automatically transcribe that spoken word, which we can use as closed caption or just embed that text into the video. But we can also export that as an SRT file, take that into something like DeepL, which I think is great on the web for translation, and then bring that back in so that we have a Japanese subtitle, and we can also have English closed caption if we want that. But let me show you how to do that right now. So I have voiceover, or in this case, monologue. So if I want to auto-generate text for that, you just go to text and auto captions. There we go. And it just takes a few moments for that, and it will put it in the correct spot. So that really is a time saver. It's not perfect, but it's 99% of the way there. We can just take the time. We can go in and change a few things, such as... Let me just speed this up here before I wrote about Presentation Zen. Okay, as you can see, Presentations Zen. That's not the title. It's Presentation Zen. And DeepL, when we get to the translation, should understand that that's a title, and we'll put kind of a quotation mark around it. But you can also change the look and feel of this text. And when you bring in the translation later, the Japanese or whatever, it'll match that. So here's what I did for this one. I just went to effects and went down here. It was this one. Okay, go back to the basic, and we can increase the size. And yeah, that's okay for this one. All right, so we got the English okay, plays all right. Even before I wrote the book Presentation Zen. Okay, so now how do we create a SRT file so that we can translate it into DeepL? So what we do is we go over here to export. And the default, actually default will probably be the video exporting. But what most people don't know, if you deselect that, then you'll see just the SRT file. And that's our captions with the time codes in them. And that's what we want. So just export that. I just put it on the desktop to keep it simple. So there it is, as you can see. Now we want to open that up into TextEdit. So I go down there, or Notepad on Windows. And then we can just select all of this English with the time codes in it. Go to DeepL, drop it in. And you can see the Japanese is there. And then we can copy that and paste. There we go. And you can see there's some things we can change. For example, it says Presentation Zen, the Chinese character for Zen. But let's say we just want to use Romaji for that, Z-E-N. And yeah, this is good to go. So we can just, it'll just save automatically back to our desktop file, this one over here. So now if we pull this in to CapCut, you're going to see there's our Japanese. We'll turn off the English so we can see it. There's the Japanese, Presentation Zen, Presentation Zen, Okakumai. So yes, that's correct so far. And probably, yeah, we're going to have to finesse. We've got some two lines of text here. But we can finesse that a little bit like this one. For example, you can go in and things up a bit. And you can, for example, you can increase the spacing on the line if you want. Whoops, make it a little bigger. That's the wrong way. So now I have Japanese. And if I want, I have English subtitles as well, depending on what I want to export. So that was really easy. And it's a big time saver, as you can imagine on a larger project. I've done this with 10-minute videos where I have quite a bit of text. And yeah, it works. 99% accurate. You just have to finesse it a little bit. Translations are really good. But still, you can't rely completely on the AI. It still needs humans to go through and edit it and design it.