Podcast Editing Tutorial (Audio-only Projects)

Today you're going to learn how to edit an audio only one person podcast in descript. Starting with importing the raw audio file into descript, I'm going to show you step by step the exact workflow that I would use to get your podcast ready for publishing. Meet me over in descript. Here I am inside of my workspace. The workspace is the view that you get when you first open up the Descript app. And in the top right you're going to see a big blue button that says plus new. Go ahead and click on that and we get a dropdown and we're going to choose Audio Project. We're doing an audio only podcast. When you click on it, it's going to open up a brand new blank project. And what I'm going to do is I need to add the raw audio file. And to do that, the simplest way is to find the file on your computer. If you're recording straight into descript, the process is going to be very similar, but your project is already going to be inside of descript, if you recorded it directly in here. So I'm going to grab that file and I'm going to drag it into my descript project. And you see it says drop anywhere to transcribe as soon as I release. That audio file has now been added to my project. And I get this box that says transcribing and I can label it with the speaker, which is just me. It's Ross and it's in English. I'm gonna hit done and now I'm gonna give this a moment while it's transcribing. And there we go. My project is now transcribed. I can see it all written out right here inside of my view. The first thing that I'm gonna do now that my project is in here is I'm gonna apply Studio sound. So I'm gonna open up my timeline. And this is our layer right here. This is our sound layer. So I'm going to click on that. And once I click on it, once I have it selected, you can see this properties menu opened up on the right side. It says script at the top left of it. And then duration. It's currently at 1x. This is where you could change the speed if you wanted to make it faster or slower. It's got the volume. It says negative three. Yours will probably say something different that's automatically set by descript. It it automatically adjusts the volume to make it a, a natural level. It automatic levels it. Sometimes it makes it plus three plus ten it's just going to depend on the recording environment. The next thing is the audio effects. And there's two things down here by default, studio sound and ducking. We're going to apply studio sound and we're going to simply toggle it on. And you can see this little wheel spinning, telling us that it's processing as well as down here, applying studio sound. And this is going to isolate our voice. It's going to take out any background noise and just generally clean up the audio to make it sound better. And he uses AI to do all that. There's nothing else required for you other than to just give it a moment. This is a six minute, 49 second podcast, so it's probably gonna take about two minutes. And while that's processing, I'm gonna show you ducking. Ducking is right below it. And what ducking does is if I apply it, I and I click on this little effect dial next to it, you can see that it says 10%. And what this does is it lowers all the other tracks to 10% of its volume. So if you have background music, it's going to automatically lower that background music to 10% of its original volume in order to give it that nice soft background while not covering up your voice. The main portion of your podcast. And it also applies to sound effects. If you have sound effects, keep in mind this will also apply to those. I don't plan to add music, so I'm going to go ahead and remove effect by clicking this little delete button. Okay, so my studio sound is done processing and let's see how it sounds. And then I can toggle it on and off while it's playing. So let me play that again. And this time I'm going to toggle studio sound on and off so I can compare. Okay. And I like it with studio sound on, so I'm going to leave that on. But if it was, if it made my voice sound a little bit robotic or flat or it was cutting off sounds, like sometimes studio sound will cut off an S sound. So like when I say microphones, it'll cut off the S at the end there. And if that's the case, I can type into this little box here where it says 100% and I could knock it down to 80% or anywhere between 0 and 100. But again, I like the way it sounds, so I'm going to leave it at 100. Now the other thing that I usually do when I'm changing the podcast sound is I'll hit this plus button and all your audio effects are gonna be in here. So you got eq, you got these different creative effects, Bitcrusher, distortion and flanger. I'm not gonna add those. Those don't sound good for a professional podcast. But the one that I do like to add is this one compressor. And you can see ducking here as well as limiter. But I'm gonna add compressor and then this effect settings. There's a default here. I usually do the one that says classic voiceover. So as before, I'm gonna play this, I'm gonna listen to it with that effect applied and I'm gonna toggle back and forth between off and on, just to listen and see if I like that effect or not. If I don't, I'll leave it off. Okay. And I like it. It boosts my voice a little bit. What the compressor does is it tightens the dynamic range of your audio so it boosts those really quiet areas and it shrinks down. It makes quieter the really high sounds. And I like it. It's a nice effect. So compressor and studio sound, those are the two effects that I usually use in my projects. And if Descripts auto leveling didn't do a good job, if it's too loud or too quiet, you can change the audio here. But when you upload it to the podcast platform, the audio is going to change a little bit again based on the platform that you upload it to. So don't worry too much about if it sounds quiet or loud, because the platform you upload it to, whether that's Spotify or Apple podcasts, whatever it might be, will auto level it on their side. That's the broad audio effects that I apply to the whole project. Now it's time to start cutting. The first thing I'm going to do is go through my whole and I'm going to do a rough cut. This means I'm going to remove all the sections that I don't want. Just the things that I know for sure is not going to be in the final product. Starting with this little gap at the beginning. Maybe you started your recorder and you didn't start speaking for five or 10 seconds. Then you can shorten that gap. So there's not such a big pause before you start speaking. This is only about a one second pause, which I'm actually gonna leave in. I'm not gonna touch this because there's no rule of thumb to this. I usually do about a 1 second to a 3 second pause before I start speaking. And the idea with that is somebody has your podcast in their podcast feed. The first podcast finishes and then it comes to your podcast. Rather than immediately jumping into talking, there's just that little pause so that they know the podcast episode has changed. So I'm gonna leave that in in my rough cut. I'm not gonna do a whole lot because I scripted this. So I was actually reading this article for the sake of this podcast, so I'm not gonna have anything to cut out. But maybe you just spoke off the cuff and you had retakes, you had parts where you repeated yourself or you went off on a tangent. What you're gonna do is find all those parts and you can actually use AI to help you identify those, which I'll show you in a second. And you're gonna simply click on them. So let's say that this second sentence here that starts with it's called let's say that I don't want that in my podcast. All I do is click on it, left click, select the whole thing and this menu pops up over it. What you're going to do is click on this button right here. It's a S with a strike through it. I'm going to click on it. And now you can see there's a line out through the whole thing. This is called the ignore function. And what it does is it still shows up in my transcript, as you can see, but it's gone from my podcast. So the, the audio portion of it is no longer in this podcast. And descript is non destructive. So at any time you can restore to another version. So I can hit undo command Z on my keyboard. I could click on this again and now it's back. And there's also a version history. If I go to file and then version history right here, I can restore to old versions of this file. I've talked about those in other videos on this channel as well as on my course. If you want to check out my course to go deeper into all of this, I highly recommend that I'll leave a link in the description. That's what I would do. Go through, find all the parts that you don't want in your podcast and strike it out using this ignore feature. Now, if you want an AI to help you, you can use this actions button at the top. This is all of the AI actions in descript. First of all, there's remove filler words. This is going to look through your whole podcast, find any time that you said, you know, things like that. It's going to identify those and it's going to help you delete them with the click of a button. So let's go ahead and do that. It found four. And if I click on this, it tells me what it found. So three times where I repeated words. One time where I said so. And so if I click on that, it found a portion. And again I was reading this. This is. This is a scripted thing. So that was an intentional sew and I'm going to leave it in there. It also looks for times where you repeated yourself because possibly you were just catching your thoughts and you said the same thing twice. In this case, it was intentional. I said. So I'm going to leave that in. If I wanted to remove any of these, I would simply click this remove button. And if I wanted to just blanket remove everything that it found, I can click remove all and it'll do that. But again, for the sake of this podcast, I'm not going to do it. I'm going to close out of that. The next one is shorten word gaps. And again I went back to actions here. This is the big gray bar at the top. I click on that, I go to shorten word gaps and it's looking for anything that is five seconds or more where there's no speaking foundation. And it found three instances of that. So let me click on this one. When I clicked on it, it's playing it. And this is a six second gap. And I do want to remove this because that was a big pause before I started reading my script. So I'm going to shorten it to one second. By default, it wants to shorten it to three seconds. That's this number in this box here. But if I click in there and I type one, then it's going to shorten it to that. And as soon as I click this word shorten, it made that change. And then it automatically moves it to the next gap that it found, which is down here. It's a six second pause. I think I was just gathering my thoughts. So I'm gonna make that change as well. I'm gonna click shorten and that's the end of the podcast. So there's a big gap at the end there as I was going to hit stop recording. And so I'm gonna hit shorten once again. And if I wanted to just shorten all of those, I knew it was a change I wanted to make. I could have hit this shorten all button and it would have done all those changes at once. And now you can see that this end here, there's just a one second gap instead of that six second gap that was originally there. So that's the shortened word gap. I'm going to go ahead and close out. I'm going to come back to AI Actions. And if there was a certain part in the script that you wanted to find, this is your search feature here. You can type in a specific word. If you know there's a section that you want to cut out, that's a quick way to find it. I'm going to close that back to AI Actions. And this bottom one that says Ask AI, this is Chat GPT. So if you click on it, there's all these presets in here and they're, they're broken down by categories. So there's edit, find, publish, and write. And if you click on one of these, this is just a prompt for Chat GPT. So if I click Remove Retakes, the prompt is detect and delete all the repeated takes from this recording. And you could, you could change that prompt. You could just completely start one from scratch and tell it anything you want. But this is the preset for Remove Retakes. And let's see if it finds anything. It probably won't because again, this was scripted. It says AI is thinking. And I just clicked on that send button to start the process. So AI is thinking. It shouldn't take too long here. And there we go. And I'll scroll through this and if it found any retakes, it's going to cross them out. And there's one. It crossed out the sentence. What did you attach to the file server will get you less information. Then what this is looking for is parts where link. Say you're running your recorder, you're doing your podcast, and you mess up a line and then you just simply repeat that line with the proper pronunciation or the proper intonation that you intended. Gonna find that bad take, identify it for you, and then if you want to make that change, you can say apply edits to script and it'll go ahead and ignore the bad take, leaving the good one. But I don't want to do that. So I'm just going to hit close and it's, it's not going to make any changes. You can comb through your entire podcast using the Ask AI. Again, that was actions down to the bottom. Ask AI. The one I just showed you is Remove Retakes. You could do edit for clarity, add chapters, et cetera. All of these things are going to help you with your podcast workflow. That would be the rough cut I'VE gone through, I've used all the AI tools, I've made a bunch of changes. The next thing I would do is play through my entire podcast and make the fine changes that I need to make this thing sound smooth and make it sound good. So I'm gonna go ahead and play through and I'll point out any areas where I have a change that I wanna make. So as I play this podcast, the you can see the active word is highlighted on the transcript and as well the playhead, this vertical blue line down here is moving. So these lines are in sync. If I click here to the word now, it moves it there on my transcript as well as it moves the playhead to the word now. And I can zoom in on my timeline by holding command and using my scroll wheel to scroll up. And that zooms in. It makes my view bigger, meaning I'm looking at a smaller piece of the podcast. You can see these numbers up here. This 1.5 is telling me where this view begins. And then this 1 minute and 5 seconds is where this view ends. So this is about a 5 second view right here. And I can scroll left and right by holding shift and scrolling up and down on my mouse wheel. But I'm going to keep zooming in. And I can see this word now right here. And let's say this is, this is a gap. You can see these sound waves with the transcribed word over it. If I want to shorten this gap, I can click on this word and I can drag it to the left. And you see that little black box with the number that's telling me how long that gap is before the next word. And now I tightened it so there's no gap at all. I'm going to go ahead and undo that though. I just did that for demonstration. And you can not only close gaps, but let's say I want to take out that word now. I could click on the word that right here and click on it and drag it over the word now to get rid of it. Now I mentioned descript is non destructive. So if I wanted to restore that beside clicking undo, I can put my mouse on the boundary where that clip was made. And you can see there's a little vertical gray line here on the transcript that's letting us know that a cut was made there as well as there's a clip or a vertical gray line right here. And if I put my mouse on that vertical gray line, if I put it slightly to the left of that gray line, I get a bracket like I Have right now, that's open to the left. If I put it slightly to the right, it's now open to the right. And that's telling me that if I click on that boundary, it's going to affect that side of the clip. So if I want to restore, I make my bracket open to the left like that. I click and I drag, and that's restoring the part of the clip that I got rid of. And if I keep restoring now, that word now comes back like so. And now my. My clip is. Is whole again. So that's the quickest way to remove things from the timeline. The quickest way in general is what I showed you the first time, where you simply click and drag to select all the words that you want to delete. So in this case, if I selected this entire paragraph and I click delete on my keyboard, it's gone. I can undo. Or the first way I showed you, which is this strikeout. Now, the advantage. It's always recommended to do the strikeout, the ignore feature, because it keeps it here on our transcript. So it just. It lets us maintain awareness that we got rid of something there in case we decide to restore it. If you delete it, you can still restore it, but you just won't see it on your transcript. So you don't have the awareness that something was originally there. I'm going to hit undo. I'm going to hit undo again. I'm going to keep hitting undo until all of my changes. Okay, now we're back to where we started. After going through my entire project, I've removed everything I want. I've gotten it all, all, all the sound sounding really good. With the studio sound and the compressor, the last thing I might do is add a little bit of music or sound effects. So right where I introduce this article and before I start reading it, like, where is it? Right, right here, I'm gonna add just a little background sound effect, just so the listener knows that I'm transitioning to the part where I'm gonna actually read this thing. So I'm going to put my playhead where I want that sound effect to be placed. So this, that's this blue line right here. I just click on it down in my timeline. And one very important thing in descript is where you're clicking, you have to make sure if you want to make a change to a layer, you need that layer selected. If I want to make a change to this layer, I need it selected, and I'll know it's selected because it's in blue. And then this Properties menu will appear. If you. If I don't want to make a change to a specific layer, I click into the white area above it because I'm going to add a new layer, which means another sound effect or music above my script layer, the layer where I'm actually speaking. So I do that by clicking into this white space right here. And wherever this blue playhead is, is where my sound effect is going to get placed. So I come up to the Media Bin at the top here, and the files is the files that you have brought in, so. Or all the files in your project. So in this case, this is the file I brought in. It's called how to Help Someone Use a Computer. If I come over to the right of the tab, the Files tab, there's videos, and these are descripts built in. Videos doesn't help us. We're not making a video. GIFs images and audio, for our purposes is the one we want. There's a Music tab, and there's a Sound Effects tab. In Sound Effects. I'm gonna find. I don't really know what I want, so I'm just gonna just play some of these by clicking on the Play button to preview them. I like that. It's a slow whoosh. It's like. And you can see on the right side how long it is. It's about three seconds. So that'll fill this space nicely. And to add it, I'll just click on it and there's a little loading media thing. And there it is. You can see it added right where my playhead was. And I can see the sound wave. And I can see the sound wave happens. It overlaps a little bit with where I start talking again. So I'm just gonna adjust this layer by clicking on it and sliding it to the left so that that sound wave happens over this empty space right here. So let me see how that sounds. And with editing, it's all about making a change, backing up a few seconds, watching it, or in this case, listening to it, see if you like the change. If not, make adjustments until you do. So let's see it. Nice. I like it. It goes from me introducing the article, little whoosh, subtle. And then it goes into me reading that article. So that's how you add another layer is by going into the Media bin, choosing the audio. You can add music, you can add sound effects, or you can add. If you have your own stuff, you can bring it off of your computer using the drag and drop method that I showed you. At the very beginning of this tutorial. So that's it. Let's say that this project is done, we're happy with how it sounds, and it's time to publish this thing. The way we do that is we come up to the publish button here in the top right, and by default it's on publish. And there's two words you need to know in descript, Export and publish. Publish is going to send it somewhere to the web, and by default it wants to put it to a web link, which is descripts web link. It'll generate a URL that you can share with somebody and then they could listen to this project. So I'm gonna go ahead and do that by clicking this publish button, and it's gonna take a moment to load. And I always recommend you start with this even if you don't plan to share that link. And if you don't, if you don't want someone to see it or listen to it, you can make it project access required using this access setting right here. Otherwise it defaults to. Anyone with the link could listen to this project when you share it with them by clicking clicking this. But the reason I recommend that is because once you publish it to the web link, it does all the processing on descript servers. And when you download this, it's going to be a lot faster. So once it's. Once you get that publish button, you can click right here, this download button, and watch how fast this is. It's going to ask me to save it or give it a name. So I'll hit save. And it's. It's already in there, so it's asking me to replace it. I'm just going to change the name slightly and then I'm hit save again. And here's the download. Look at that. Took three seconds, took two seconds. Basically download complete. And then I can open it or I can just close this little toast and it's gone. So that's the easiest way. That's what I recommend. The other thing is, if you do export, you can export. Make sure this says current composition. If it says current scene or current selection, it's only going to grab a portion of of this video. So do current composition, that'll be the entire project. And then you can choose your. If you want AAC, type of file, WAV or MP3. Most podcast platforms prefer MP3. So I recommend that hit export and it's going to save it to your hard drive. But back to the publish for a second. If you click on this web link. You have other options here as well. So if you use any of these as your podcast host, Blueberry, Buzzsprout, transistor, et cetera, those are, those are some common ones. Then you can click on those and you can hit publish. What this is going to do is it's going to open up your browser, you're going to have to log in and give descript access to be able to communicate with your podcast platform. And then you can publish this directly from descript to your podcast platform. If your podcast platform isn't listed here, then you'll need to do the export method that I first showed you. So you'll hit export, you'll put it on your computer and then. Oh, see how much slower this is than the, than the first publishing method I showed you. I'm going to cancel that, the publishing method here where I did the download. So let me do this one more time. Hit save. And look at that took a fraction of a second. I'm going to hit open and there it is on my computer. So then I would go into something like Anchor, which, which is the podcast platform that I use, and I would drag this file into Anchor and it would then be published to my podcast feed. This has been my complete workflow for taking an audio only one person podcast, cleaning it up in descript and publishing it to my podcast platform. If you like this or you want to go deeper, check out the Descript Mastery course. I'll leave a link in the description.