Two-person Audio-only Workflow

Test, test. Okay, we're live. It always messes me up when I can't hear myself. Going to get used to this someday. So today is going to be all about a two person podcast audio only workflow. And I'm going to walk you through the process that I would use to edit this thing from start to finish. Now the first thing is this came from. These are the files I'm going to use. And this came from my lavalier microphones. So there are these little things right here, these little DJI microphones and they do a pretty good job. They clip onto your shirt and they record audio. That's it. And so but the interesting thing about this is it splits it up. So you can see here that this is two people talking. It's Bruce and me, Ross. And I had to. What these microphones do is every like 20 minutes or so, maybe it's every 30 minutes, they create a new file. So this was one continuous two hour conversation, but it split it up into four different files per speaker. So the prep work that I mentioned is to go through, label these things how you want if, if your microphone does that. So I just point that out because there's a little bit of legwork that you'll have to do depending on how your project was recorded. And you know, I heard that with like documentaries, for example, like Netflix documentaries, they just have, you know, tons of cameras, tons of files, hours, thousands of hours of footage. And they got to sort through all that. So there's people whose entire job is to just go through, organize things, tag things, and then, you know, put things into folders. So then it's a lot faster on the back end to just go through and be like, okay, here's the folder about interviews with the subject. Here's the B roll of nature folder, whatever. They just get that all tagged and polished up and organized before they start doing the actual editing. Okay. And this is obviously really simple. It's just eight files total. But just thought I would mention that. So the first thing we're going to do is come to the new button. And as I said, this is going to be an audio only project. So I'll go to audio project and I'll call this interview. I'll hit enter and then I'm going to come up to my media bin. I could drag all of my files into there, into just directly into my project. I'm going to put them into the media bin first. And this is another step of organizing everything. So I'm just going to highlight them all and Drag them in like so. And just checking the chat and everything. Okay. And I'm just going to give this a moment to add those files. It's uploading them all and you can see how they're all 31 minutes, 4 seconds. That has to do with the cutoff of my microphone. And then that, that one that is like a different amount of time. 29, 23, 29, 27. Those must be the last ones when I hit stop recording on the microphone. So what we're going to do and what this just did a moment ago is it default selected all of them, but I'm going to highlight all these. So I just clicked on one held shift and then clicked on the top one to highlight all of them. And what we're going to have to do is make a sequence out of this. Now sequences are one of the most complicated yet essential skills in descript. Sequences are how we do what I'm about to show you with more than one speaker. It's a container of all of your files to keep everything in sync. So that if you make a change to one of your speaker's files, like let's say I cut out a portion where I have a filler word, I'm saying a lot and I cut those out. If I only cut my track without cutting the other speaker or speakers tracks, then those tracks would remain how they originally were and the whole thing would get out of sync. Everything beyond the point where that cut was make made would be out of sync. We, we don't want that. So sequences are how we keep everything together and it allows us to edit everything at one time. So with all those selected, we're going to hit create sequence and these things are still loading. So it's kind of a little, it's running a little bit slow. But bear with me here. And this is going to be a little challenging because I labeled these as just the order that they were in. So like Bruce1, Ross1 and etc like that, just in the order that they happen chronologically. So what I'm gonna have to do is space these out and you can see the sound waves starting to come in. These are still processing, getting uploaded to the cloud, getting uploaded to the project. So right now we have eight different layers. And what I'm going to actually do to simplify this. Oh, and that's a max. You can only have eight tracks. So please delete one before adding a new one. What I'm going to do to simplify this is I have my wave one down here and you can See, they're labeled here as well as on the file itself. And these are all dot WAV files. So. So assuming that in my prep work I labeled these correctly and this is in fact the first clip, I'm just going to pull my second one and maybe I can't even do that. Hang on. I might have to add it. And everything's running really slow. One second here. So one way to do it is I can have these staggered like this. So I have my first ros one, the very first clip I have, and then it runs until ROS 2 starts, and then that's going to run until ROS 3. And again, if you're just joining. The reason that I have to do this, the reason I have separate files is because my particular microphone, my little lavalier microphone, chops the video up. So ideally you have a microphone or maybe you recorded directly into Descript or Squad Cast. That would be easiest of all. And you wouldn't have these separate files like this. But I do. So I have to do this. Let me reposition these so it's kind of like a nice waterfall like that. And then I'm gonna do the same thing for my guest. And I'm gonna get these more synced up in a moment. I'm gonna make this more efficient. Okay, so what I just did is I cut one of these and I pasted it. So I'm just trying to simplify, get everything onto the same track. So I'm gonna right click, I'm gonna cut it and I'm gonna paste it down onto my main track. And this will get all of the speaker named Bruce onto one line. Cut and paste. Whoops. Hang on. That's not good. Do you see that? I cut off the. I cut too early. So what I want to do is cut after down here. There we go. Okay. So that's gonna be my Bruce master track. And then I'm gonna do the same thing with my own portion. And let's zoom in a bit here. I'm not gonna make them super tight just yet because I want to. I need to zoom in a bit more to get more detail. Make sure I'm doing it. Positioning these correctly. And I lost my. There it is. Okay. And I apologize. It's running very slow, so it's kind of hard to. It's very finicky. Let me see if I can close some things on my computer. Can't close the browser. Close quicktime. Okay, let me re. Attack. All right, got that. I'm going to take my third track. Cut it paste. Fourth track. Cut it paste. Okay, so all that annoying work is done. And I'm just going to reposition these so they're a little easier to see. So put Bruce and Ross at the top. Okay, very good. Now I'm going to zoom in, and I can start to see what's going on with these tracks. And this is going to be a challenge to sync. This will be good. What I'm going to do is. Let's see here. I'm going to go back to the sequence. Going to. I'm just going to start playing it. I'm just going to get a rough idea that I have this all in the right order. So zoom out. And you can zoom out with shift or, sorry, command while scrolling. So hold the command button on your keyboard and use. Dang. It keeps going all the way back. There we go. Okay, I am now at the beginning. All right. This is gonna be really tricky to sync up. I picked a. I picked a challenging one. Maybe what I should do is just start with one track each and then let me create. Okay. I'm going to start a different sequence with just one track, the first track for each of us. So I'm going to say create sequence, add media. And all my media is already here, all those files that I already added. So I'm going to do Ross one. And then on the second track, I'm going to do add media, and I'm going to do Bruce one. Add new track. Okay. And I might have. It's possible I messed up the. The order when I was labeling these, because when I just previewed it just now, it sounded like it was off. But let's see. Okay, now those are correct. So what I'm going to do is just reposition these. And what I'm looking for is some sound waves to match up. So, like right there, I see some peaks. Those probably go together. I see a peak up here, a peak down here. So I'm using the sound waves. And you can see they look pretty. They should look like mirror images of each other if you're in the same room. If it's like a zoom call or something, obviously the sound files are gonna be unique to the speaker. And then in that case, it's going to be like, let's say I'm speaking, so you'll see my sound waves. And then Bruce responds. So you'd see his sound waves after mine. And the other challenge here, I said I picked a really hard one. This was recorded in a car, so the audio is not good. It's hard to hear. And what I'm going to go ahead and do is apply studio sound to both of these layers. And again, this is the sequence view. So I'm doing that inside of the sequence view. And when I repositioned Bruce's track, I just dragged it left and right. You can just drag it to reposition. And it would help if I had some sort of reference point like I mentioned, matching up these sound waves. If I had like a clapboard like they use in Hollywood, or I'd snap my fingers or clapped, then those would create spikes in the sound waves on both microphones where I could match those spikes to each other. So that's probably might be what I have there. And let me mute my microphone when I play because I know that causes an echo. Okay, I'm going to mute myself and I'm going to just listen to this for a bit to try to sync these tracks up. And then I'll explain what I did.

I need to be in the United States, or I want to be in the United States to attend the weddings of my two nieces. And so. And so then that puts an end point.

Whoops. I was still on mute. So what I need to do is match up the sound waves. So, like right here where he starts talking. And I can get more precise the more I zoom in. So as I zoom in, I can. From a distance, it looks like those are pretty good. Like they're pretty synced up. But as I zoom in, I can see that's not the case. This track is actually a little bit earlier. And so I'll just move that on top of it and let me just scroll. I'm just gonna scrub through this quickly and see if I can find any other peaks to match up. Let's look right here. See, I got a peak there and a peak there. Yeah. And some of those guys are still alive. Okay. I think that's pretty good. I'll hear once I dig more into these tracks. And it says studio sound is applied, but it doesn't sound like it. There we go. That sounds much better. So for some reason, this happens sometimes where it says it's applied, but it's not. So just. You can just toggle it off and back on. And. And you can see the sound waves changed. Look, there's. With it off, you see all those sound waves for the background noise. And then when I put studio sound on, it knocks all that crap out. And. And this is actually a lot easier. I should have started with that because once I have those Voices isolated. I can see that it's, it's synced much easier without all that background noise. So that looks a lot better. And I'll just zoom in. I'm checking the sound waves and I'm just gonna nudge it a little bit. Okay, so that's it. That I haven't cut anything. I just synced them up using the sound waves. And then once I'm done with that process. Well, first of all, let me finish up here on the sequence. So you can click each of these individual tracks. So there's Bruce and there's Ross. And if I anything I do to to Ross track like you saw me toggle on and off studio sound, it only applies to my track. So if I had like a better quality microphone or my sound was just better, then maybe I don't apply studio sound to myself, but I do apply it to my guests. So you can granularly control that here as well as, let's say I'm pretty quiet, but he's really loud. And actually I can tell D descript, when I imported it, automatically adjusts the volume level and it made mine positive 9.7 decibels while it made his. If I click on his and I look at the audio level, it's positive 4. So descript recognized that on average my microphone was quieter and so it boosted my volume more than his. So that's it. That's it for now. I'm going to go ahead and hit done. And then what I'm going to do is insert this into my script. So I got that sequence selected. That's the one that I just created. I'm going to hit insert into script and it's going to transcribe it and it automatically asked me. So again, These are the two files. There's Ross 1 Wave, Bruce 1 WAV. There's one speaker per audio file and it's going to be Ross 1 and Bruce 1. That's fine. I'm gonna hit done. And that's for the purposes of transcription, which it's going to start right now. So this is probably going to take a few minutes. They're two 31 minute files and I mostly I 95% of the time with Descript, I'm editing videos. So if you're, if you've watched my channel for a while, then this probably looks weird if you've never done audio only editing. And there's a few differences I'll point out, but the biggest one of course being there's no canvas, there's no video preview because there is no video. So, yeah, I'll just point out some differences as we go. One thing you can do if you go to file, so you go to the descript thing at the top, the descript logo, and then you go to file. You can switch it to a video composition at any point. And I can do it right now. It happens instantly. So there's that video file. And then I could, if this was a voiceover or whatever, I just wanted to add B, roll over it, or an audiogram, then I could do that. But I'm going to switch it back to audio only because that is the purpose of this particular one. So the biggest thing with audio only editing is there are no scenes. That's like a big thing that. That always throws me off when I switch to audio only is if I hit slash either in my transcript or here in my timeline, there's no scenes. And so basically you're. You've got to split clips if you want to make fine adjustments to the audio. And I'm going to go ahead and start listening to this and apologize. It's going to be, like, less viewer friendly, but I'll try to, as much as possible, chime in with my. My thinking and what I'm doing, explain what I'm doing as I go. And as always, feel free to leave comments and questions in the chat and I'll get to those. We don't have to just talk about this. Press one button on there. It's going to record on there as well. Okay. Yeah, Cool. And so Bolivia is high on your list for next places? Possibly.

Well, you know, here.

This is gonna be tedious to unmute, mute myself and unmute myself. So well, first of all, you can see the transcription just kicked in. So my transcription is blue, dark blue, black. On the timeline down here, the sound waves. And the red pink one is my guest, that's Bruce. And so that's one quick way to tell what's going on visually as well as it got labeled here. So there's Bruce one, and then there's me. So what I can do, what I can see is that right off the bat, there's these dots here. And those are just blank space before we start talking. So I can select those, I can hit delete, and they're gone. Now, if I wasn't in a sequence, this, this view right here, this is the sequence where there's multiple speakers going on at once. If I wasn't in a sequence and I cut either my track or my guest track without cutting the other they would now be out of sync for the entirety of the video and that or for the podcast here. And that would be a huge problem. Same thing with, with filler word removal. If I did filler word removal without this being in a sequence, I'd run into big problems. So what I'm going to do, though, I'm going to go ahead and do that right now. Go to actions. Go to remove filler words. And it found 276 in this half an hour segment with two speakers. And. And if I click on this, this is what it's looking for. Repeated words. It found 126. I might leave those in because it's just a conversation. And then the ums us, those can pretty much always go. The only time I would leave those in is if it clips the sound and it's warning me. Down here, sections of Ross 1 and Bruce 1 have duplicate transcription, perhaps caused by mic bleed. So I can show you that in a moment. But this is what it was looking for and it found 51 ums, 43 us, so et cetera. And I can unselect things if I don't want it to delete those. And then by default it wants to delete it, which, which means it'll remove it from the transcript and it will remove it from the video. I always recommend people do. Ignore. Ignore. Will leave it in your transcript, but with a strikethrough, and it'll still remove it from your video. So it allows you to maintain awareness that something was cut there in case you want to restore it, then you easily can. Yeah, that's fine. I was just deliberating whether to do the, the repeated words. So then what I can do is I could remove one at a time or I can remove all, which I'm going to do because I don't, I don't want to bore you. So remove all. And let's just give it a moment. Ignore or. Yeah, it ignored 208 selections. And so now you can see all these chops, all these gray lines in my timeline. Those are all the points where I got cut 208 times, to be exact. And then here on the timeline is what I was talking about. It remains there with a strikethrough. And as I'm playing through this and I'm listening to see if, if I like the changes that were made, I can hover my mouse over it and I can hit this little back arrow and to restore that, that filler word in this case. And now it's back just like that. But that's probably a good edit because it's an followed by a space. So I'm going to hit undo to bring it back. And now it's back. And then there's other ways to restore it as well, which I can show you as we go. I'm going to bring up my timeline so it's a little bigger. And as with always, as with editing video or any other thing in descript, you can add multiple layers. Right now, I only have a script layer. This is the layer that was transcribed. But I could add sound effects, I could add background music, I could add anything I want. Above my script layer here. And on my script layer, you can see the words. So these are the words that got transcribed from each speaker. And there's a number of ways that I can edit these up from the timeline. I could click on a word like, and for example, and I can drag it, and that's creating a gap right here. So it's just gonna be a silent space. Or I can drag it to the left and that's gonna close that gap all the way to where I could cover up previous words. So I'm removing the words in front of that particular word. And right there. Now, the other thing is. So you can see when I put my mouse in here, it's highlighting just a certain portion in blue. That is a clip that I've isolated. When the, when I remove the filler words, it automatically made cuts here and here. So those are those gray lines on the timeline. And if I put my mouse on it, on those boundaries, you can see it turns into a bracket. Right now it's a bracket open to the left. So that's telling me that whatever I'm about to do, if I, if I change this boundary, it's going to affect that clip to the left. So if I click and I slide to the right, it's restoring parts of the track to the right. And look, it's restoring stuff that's already in there. So now, now it would play that portion twice. But before originally, it just got rid of the word. I'm going to undo that. And if I do the same thing, I got my bracket open to the left. I. I can click and I can drag to the left. And that got rid of the word and right there, undo that. Um, okay, so I showed you the clicking on the words, I showed you the boundaries. The other way would be to just do it from the. The transcript itself. So if I click on the word and up here and I highlight that, I can just hit Delete and it's gone. But I'm going to undo that and you can see. So I have it highlighted on my transcript. It's also highlighted in blue down here. So a lot of visual clues to tell what you're looking at, what you're manipulating. And additionally there's this duration thing right here. You could change the playback speed of that clip that I have selected. And it's telling me that that's a 1.1 second little clip right there. So lots of different clues and context because sometimes things happen and you're not sure why it's behaving in a particular way. And if you, if you just taken all the different context clues then you can make sense of like why, why when I drag this thing did it change the different clip and etc. So, so yeah, so there's my bracket open to the left. If I move it just a tiny bit to the right, now it's open to the right, telling me that it's going to affect that clip. And so if I click and drag to the left, it's going to extend that clip to the right. And if I do the opposite, it's going to shorten that clip to the right. And that was the word so which is probably just a filler word. So I'm going to go ahead and leave that out, remove that. Actually, you know what, Sorry, sorry to confuse. I'm going to bring it back and I'm going to find it on my timeline. So this is a common question I get too is you see this blue playhead right here when I click somewhere. So right now in this case on the word so I can also find it in my transcript. So whenever I move the cursor down here, like I just moved it to the word around and you can see my transcript just scrolled up and it put the blue playhead here and here. So back to the word. So click over here. It scrolled my transcript down. My cursor is now in front of the word so. And I forget what I was going to do here. Oh yeah. So if I highlight that word, it's highlighted on, on the transcript, it's highlighted on the timeline. I can hit this thing right here, this S with a line through it and that's going to do the strikethrough, the ignore feature. So it's now gone from the timeline, but it's still there on the transcript. But it's slightly, it's gotten lighter, it's a light gray with a strikethrough. So I know that I removed something there and here as I. I'm just gonna, you know, again, feel free to leave questions in the chat. I'm just gonna mention things as I see them. So right here I can. Looking at the sound waves, I can see that it kind of abruptly starts with this word if. And let me just hear how that sounds.

If I've got. Let's see.

Okay, that's not really a good example of what I was looking for. I was looking for a clipped word where the word starts too abruptly. And if I find one, like here, maybe I don't want to go all. So I can see, looking at these sound waves, that it's cut off a little bit. Like, you see, if I extend it, that blob keeps going until it tapers off right here. So if it's cut off abruptly like this, then it's probably worth first of all listening to it, see how it sounds. If it sounds terrible, then extend it using the bracket to restore until you get this taper. Because those are the natural pauses in speech between words and sentences. And so that's how you. That's how you can smooth things out. And so again, a reason why I use ignore instead of delete, because now I can see what was removed in the transcript, and then I can restore it to make it sound a lot more smooth, if that's what is required to make a good edit there. Good editing is subtle. It shouldn't be jumpy, it shouldn't be obvious. And like right here, this is probably a problem. You see how the sound waves are getting cut off?

It takes time.

Yeah, there's a weird sound there. So I'm just going to cover that up. I'm just going to get rid of it. Put my bracket on the line, slid it to the left, and it's gone. And there's some parts, like, you can see these black parts that have no words above them. Something happened there with the transcription. Either it didn't detect what that word was, or it might just be a noise. It could be like, you know, like I said, we were in the car, so it could be honking. It could be something going on. That's not a. Not a word, but a lot of times it's just that it makes a mistake. It didn't recognize it, that type of thing. Okay, so here is some of. Here I'm speaking a bit more. It's blue starting at this point, and you can see the red waves inside of that. That could be him saying, you know, nodding his head, saying, huh, huh. It could be something like that. It could be crosstalk, like someone talking, trying to talk over me. Like probably not here if I had to. Like possibly here where the sound waves between the two colors are equal. This could be us talking over each other. And let me just play it and then I'm going to show you how to fix that.

Yeah, whatever's bordering in nearly every country.

Yeah, so that was us talking over each other. So what I just did is I right click on it, I hit edit sequence and I get back into my sequence view and you can see here my track, the blue one, and then his track. So what I can do to edit this and this doesn't always work how it should. So let me. Let's see if this will do it. I use the range tool here. This is at the top. This allows you to select a specific starting point and a specific ending point and everything in between. So I just selected all that part where my sound waves are crossing over his sound waves. And it could also be micbleed where his voice is getting picked up by my microphone. So it's getting picked up twice. His microphone and mine, which you don't want either. But this is how to. How to cancel that out. You select it and then if it's working, it's not going to work. I just muted it. But what it should do is mute just what I have selected. But for some reason, ever since a few versions ago in descript, this feature doesn't work and it mutes the whole track. So I'm going to undo that. So I'm re adding the sound back. So what I do instead is right here. I'm just going to hit delete and let's give it a second. I hit delete on my keyboard, but it's thinking about it. It's being really slow. Come on now. Oops. Not sure why it's not deleting. That's going to remove the whole thing. Dang it. It's being really slow. Sorry. Okay, see what I just did? I won't. I won't even say it. I'll just confuse. Okay, you know what I'm gonna have to do? This is the problem. This is. This has been not working for a while. I'm gonna have to use the blade tool and actually it's fine now it already because I moved it a little bit. You can see a little bit sliver right there and a little sliver at the end. So I isolated this clip. So if I click on it, it's just that portion. I'm gonna Hit delete now. Oh my God, it's still not working. Okay, well, I'm gonna mute this. I'm not giving a very good demo here. So I just muted that portion that I have selected without affecting the rest of the clip. Although I do see that it lowered my volume, which when I did the mute and unmute. So I'm gonna have to fix that in a moment. But that part, that crosstalk has now been muted. So when we play it, you're only gonna hear his track and Brazil and.

Yeah, I think it borders nearly every country in South America. I mean, at some.

So that sounds much better. It's not two of us talking at once where you can't understand either of us. It's putting the focus on his track. And you could scroll through and just visually, very quickly, visually, like right here, this is definitely, definitely mic bleed. He's saying, I'd prefer to stick closer, maybe to Nicaragua. And then the same thing is transcribed on my track. So let me show you this again. I'm switching into the blade tool. It's this little razor blade looking thing. I'm going to select a portion after where that mic bleed is happening, any portion before it to just split it. I'm going to switch back to select tool A on my keyboard and I'm going to select that. And now I just have that part that I isolated selected and I'm going to mute it. And there we go. And now it's just going to be him talking, which is what we want. So. And since this is mostly him talking and it's just me occasionally asking questions, what I might do in this case is have me muted by default to help fix up that, that mic bleed and the crosstalk. And I can also tell visually looking at this, his sound waves are much bigger than mine. So there's a couple things we can do. If I right click on this auto level clip that was. Right click auto level clip and this will apply the auto leveling that is done. And you can see it's much bigger now. It changed it to positive 9.6, but it's only going to be the part that I have selected in blue. So I'd have to go and do that for this other part and this part as well. And then those parts will be auto leveled based on what descript. And you can see this. This section it did as a positive 11. This little section it did as a positive 9.6 as well. So it just depends on the sound within that selection for how loud or quiet it's Going to make it. Okay, so we've got studio sound applied. You know what I'm going to also do is I'm going to select this whole thing and I hit. I selected the first one and I selected the last one while holding shift to select multiple tracks at once, multiple clips. And I'm going to apply compressor to this. So I go to Audio Effects plus Dynamics Compressor and I'm going to hit these little effect settings here and I'm going to go to Classic voiceover and let's see how this sounds.

And once you're there, you might as well enjoy it. Spend time there.

And Chile is so thin, it's probably super easy to just cross into, like anything. So my voice sounds pretty good. He probably needs it as well. I'm gonna do Audio Effects, Dynamics Compressor effect settings. And these are just defaults. I don't really. You can play with these. You can change the threshold, the ratio, attack, release and knee. I find that the default works pretty well and saves a lot of time on the way. Like, which would be what, Argentina and Brazil.

And yeah, I think it borders nearly every country in South America.

Sounds pretty good. There's definitely parts that are going to be a problem in this track where the studio sound is a little bit too powerful and it chops our voice. So in that case, if I select a track like mine, I can go to the effect settings, click on that, and the intensity is 100% by default. But. But I can make that number anywhere between zero, which is equivalent to not having studio sound applied at all, all the way up to 100% and so it'll blend that somewhere. And so later on when I'm. I'm fine tuning this, I'll dig into that and probably end up putting it somewhere around like 80%, 90%, something like that, just so it's not making a sound too robotic and not chopping off any sounds. So, yeah, that's. That's a lot that we covered there. Let me find another part with crosstalk just to show you a couple more visual things. So. So here we go where there's some pink on top of the blue. So I can tell that. And some of those guys are still alive. So that's probably just Mike Bleed right there because he's not saying anything. So go. Edit sequence. Yeah, it's definitely Mike Bleed and those guys alive, which is the same thing I was saying. So I'm just going to isolate that, clipping it before and after back to my select tool. I click on it and I'm Going to mute it. And then. So. So, yeah, it's fixed in the sequence I had done. And you can see that the overlapping tracks, the pink that was on top of the blue is now gone as well. So another visual cue to let me know that that has been resolved. And there's a lot of crosstalk and things happening. A lot of overlapping colors going on here. And you can see how this would get crazy, too. The more tracks you have and the principles are the same. You go to edit sequence, and then you could have track three, track four, track five, all the way to track eight. And then you'd have a bunch of colors. You'd have green and yellow and purple, and you'd have a lot going on. So it gets exponentially more complex the more people you have. Not exponentially, but it adds complexity, obviously, the more speakers that you have. And then, of course, if you're doing this with video, if you have a cut, these might all be jump cuts. So that's the good thing about audio. Audio is more forgiving in that sense, where you can't see what the speaker is doing if at one point he's leaning into the camera, and then after an, he's way back here. Audio is much more forgiving in that sense. But the principles are the same where through editing, you just kind of. You make a change, you listen to it. If you like the change, you keep it. If not, you keep adjusting until it sounds smooth, until the edit is subtle and you can't even really tell that it happened here. Looks a little choppy. Other body language.

Yeah, it was really fun. I enjoyed.

Yeah, that gets a little cut off there. So if I just extended a little tiny bit, I can just get in that. That purple sound blob to right where it ends. And that's a way to just make sure that it's not prematurely clipping anything. Other body language, right?

Yeah, it was really fun.

And I could also go into the sequence and mute him out there. That would be another option. Let's see. Trying to think. Yeah, so the first pass would be. And Descript makes this easy, is the rough cut of just editing for content. Like, okay, what tracks am I going to actually include in this video? And so that would be the first thing. And then you can use these Ask AI features, which I went into at length in yesterday's livestream. And you can go through these, like, edit for clarity will do a lot of things. This is the prompt. Edit this down to remove any speech disfluencies, repeated sections and tangents that Distract from the core ideas. So if I run this, probably going to find a lot because this was a wide ranging, free ranging conversation. AI is thinking. And those filler words, those are probably the ones that have already been removed. But since they're still in my transcript, they're still showing up here. Here's the thing where I was explaining about the show and I couldn't remember what country. Like Bolivian. Bolivian, Uruguayan. And so I think it found those and removed those things like that where it'll identify either retakes or parts that are like that. That's a perfect example where I was thinking about it and it chose the correct one. So yeah, I could, I could dig into that whole thing and then I could say apply edits to script or I can close it out, which I'm going to do and then ask AI. You can apply any of these other things. Add chapters or markers is super useful. Let me, let me, me add like 12 chapters. Hit enter. AI is thinking and I'm going hit add to script. So chap markers. I've gone into these at length in the last few days, but there you can see this here bolding. And this was AI generated starting the adventure dream versus reality. Once I add markers, this button right here appears. And since we don't have scenes on the left side in a video, we would have scenes right here where we could navigate through the markers and the scenes. But in lack of scenes, markers have a different kind of function with audio only. And so you can quickly scroll through here and be like, you know, these are labeled adventures and hostile life in San Gil. And it takes me right to that spot in the podcast where he was telling a story about that. And so if he like, well, you could use this in infinite ways. If he was like, oh, could you please remove that story where I talked about San Gil? And then I could find it, I could highlight that whole portion and delete it. Or if nothing else, if I was like, oh, that was a really cool story. I want to make a clip or a teaser out of that. I can navigate straight here and then I could highlight a portion, I could copy it and then paste it into the beginning or however I plan to use it, maybe I make it into a audiogram to use on a social media platform as a video. All sorts of useful ways that these chapters come into play. Oh, and also if you see we can't put it to YouTube. So what I could do if I go back to ask AI and I do like a. I could say like show notes, draft podcast Style show notes for this I hit enter. It's going to generate based on this script show notes and let's see if it comes up with timestamps and everything. It should. Yeah, so it took my markers and it made them timestamps where each, each turn of topic occurs. And then when I upload this to Spotify, Apple Podcasts, whatever my podcast platform is, if I use this then this would be my. That would go into the show notes and then the podcast player would automatically pick up those timestamps and divide it up into those nice little chapters. So all super, super, super handy tools that take away a lot of tedious manual labor. And then at the top we have like a suggested title. Global Nomad Adventures and Language Learning. And then I could read through this if I like it. All I can do here is copy to clipboard. This does not put it into the script and it'll just put into my clipboard and then I could paste that into my podcast platform of choice. I use anchor for my personal podcast, but whatever you use, transistor, buzzsprout, anything like that, that is where you would put it. And then what else could we do with AI actions? And again, feel free to. Haven't gotten any questions yet. Feel free to drop things in the, in the chat. We can take this in any direction. Add scenes by active speaker. Doesn't make sense. It's not a video. Good clips. This could be interesting if I want to pull out some things here to make into videos and put onto social media. I could do that with this prompt. AI is thinking and there it is. So if I hit select in script, these are the things that it, it would identified as standalone stories that can make good clips for something like YouTube shorts or TikTok back to Thailand. So yeah, a lot you can do with that. This really, really streamlines the process of the post production process of like making these show notes and adding the timestamps and finding the social media clips, all that kind of stuff. This really helps with that. Okay, we're down to our last five minutes and I haven't gotten any questions in the chat. So what I'm going to do is tell you about my new free course which is this thing right here and it's you sign up for it and every day for seven days you get an email. It's a drip of different skills and lessons as well as an assigned project. So throughout the course you are working on a project in order to practice the skills that you learn along the way until by day seven you're complete with the project and you publish it and export it. And you can just go to descriptmastery.gumroad.com where you can sign up, just put a zero in there and then hit I want this. And the emails will start right away and then you get a new one every 24 hours after that. And the link to that is in the description. So yeah, check it out. And if you don't want to do that, you can see my website, descriptmastery.com Oops. And I went to the didn't change the end there. And this is just the website where you can find tutoring, you can find the full course, and you can find templates. I create templates for people, for businesses, as well as a bunch of free downloadable ones, including a fun one I did today. I did a challenge for a podcast called My First Million and I created a cool template for that which I'll make available on my my free downloadable store here. But that's it. Thanks for watching and last call for questions. I'll be back tomorrow, 1:30pm 1:30pm Mountain Time. That is 12:30pm Pacific.