One of last year's many disruptions was in the way we consume content. There was a huge boom in audio, as people embraced podcasts, audio books, and snippets through devices like Alexa. This is an important development for business owners. By creating audio content consistently, you can build authority and trust with your audience. Being in someone's ear creates intimacy. Your voice can offer reassurance, inspiration, or much-needed levity.
Editing audio can be fiddly, though. Either you need a degree in sound engineering, some very expensive software, or spare cash to hire someone. Descript is an exciting new solution, offering a simple way to edit both audio and video. In this post, I'll just focus on audio and show you how to easily tidy up a recording.
With Descript, editing audio is as easy as editing a Word document. You import your file, wait a few moments for the app to transcribe it, then you can start tweaking the transcript. If you delete a sentence in your transcript, it's also deleted from the audio file itself. You can also rejig the order. You might've noticed that podcasts often start with a snippet from later on in the recording. This can be done simply by copying and pasting that content in Descript.
You can also perform a range of tidying tasks in Descript, such as removing umms and ahhs, reducing the length of gaps, and fixing bloopers. Here's how you can record and clean up an audio file.
1. Record yourself using a good quality app (it should be in .WAV format, which is higher quality than MP3). If you don't already have a favourite, I'd recommend Easy Voice Recorder for smartphones and Audacity for computers. Here's the important part: as you're recording, flag any bloopers with a distinctive word (it needs to be in the dictionary, so inventive swearing won't work). Or you can just calmly say "Take Two". Whatever you choose, it must be both distinctive and consistent. I'll explain why in a moment. There's no need to start all over again if you mess up - just flag it and continue.
2. In Descript, click Add new, then Composition, and give it a name. Either drag and drop your WAV file, or click to locate it on your computer. You'll see a message to say that Descript is Adding file.
3. When the file has been added, you're invited to click Transcribe in the bottom left-hand corner. This might take a few minutes, depending on the speed of your computer and internet connection. While Descript is pretty accurate, it's not 100%. Don't be distracted by any errors at this stage, though. You don't want to spend time tidying up content that you'll then delete.
4. Make a copy. Click the ellipsis next to your composition and choose Duplicate. This is so you can easily revert to an unedited version. While you're getting used to Descript, you might get carried away and end up with a big jumble.
5. Now it's time to find those bloopers. Using the search box in the top left corner, enter the word or phrase you used to flag those errors. Assuming Descript transcribed them correctly, you'll then see them all in a list. You can click to hop directly to each of them.
6. Once you've located that mistake, you can select and delete it - just like that! You'll notice the that the waveform (the visual representation of the sound file) at the bottom is changing as you edit the transcript.
7. Next, find gaps between words. Click the tiny spanner in the toolbar and choose Shorten word gaps. You can then choose the maximum length for those gaps and Descript sprints through and makes the necessary changes. Silence is much more noticeable in audio than in video, as there's nothing else to keep your audience engaged. If the gap is any longer than a couple of seconds, they might think you've finished. Dramatic pauses should be used with caution. Experiment to see what sounds right. Press Tab to start and stop playback.
8. Descript can identify common filler words, such as um and ah. Click the ellipsis next to your composition and choose Remove filler words. The Pro version also picks up other horrors such as like, you know, and sort of. Unless it's a long recording, it should be easy enough for you to spot them yourself. Select them in the transcript, then hit delete. You can use this technique to remove any other oddities. If any random gaps have emerged, use the Shorten word gaps tool again, or select them in the waveform editor by clicking and dragging your mouse, then pressing delete. Your audio file should now be in much better shape.
9. It's good practice to include a transcript with audio content. This is mainly for accessibility reasons, but some people find it faster and easier to read than to listen. Switch to Correct Text in the top left hand corner. This is very important, otherwise any changes you make affect the audio file, rather than just the transcript.
Once you're in text mode, you can start correcting. If it doesn't look right and you're struggling to remember what you actually said, place your cursor by the text and press the tab key to hear the audio. You can change the playback speed and also get it to automatically pause when you start typing to make a correction.
10. With an edited audio file and a tweaked transcript, you're ready to publish it. Click Share in the top right hand corner, and then Export. For the transcript, choose Text; for the audio file itself, select Audio and then the format you require. If you're going to do more editing in another application, such as Adobe Audition or Audacity, export it as a WAV file. If it's ready to upload, go for MP3, as this is a much smaller file size.
One limitation with Descript is that it doesn't currently handle noise reduction. Recordings often feature an annoying background hum that comes from computer equipment and radio interference. The best workarounds are to either record on your smartphone and make sure all your other equipment is switched off, or to use noise reduction tools in another application. The simplest solution is Audacity, free software that works on PCs, Macs, and Linux. I'll show you how to use this feature in my next tutorial.
The good news is that Descript is working hard on adding noise reduction. Their developers are highly responsive to user requests and very speedy on rolling out improvements. This limitation aside, I think Descript is a great tool for editing audio without too much faffing. It's ideal if you're getting started with podcasting, or just want to make the occasional recording. And Descript works with video, too, so it'll help you with future vlogging adventures. More of that in another post.
The free version of Descript offers three hours' transcription, which is definitely enough to get you started. Once you've got the hang of it, the Creator level gives you 10 hours' transcription each month for $12. The Pro level at $24 includes 30 hours' transcription, along with more sophisticated tools, including an overdub feature. With this wizardry, you can correct your speech in the audio file, just by typing.
Anyway, muck around in the free version to get started. To share your work, you can host 3 hours of audio for free on Soundcloud. Then you'll be on you're way to reaching the ears and minds of your audience.
I don't have any affiliation with Descript. I just think it's a smashing piece of software and helpful for my readers.