This script was created by otter.ai, an automatic transcription service. The original audio recording of this article was created by Bodie Grimm, creator of the Kilowatt Podcast at 928digital.com.
To hear the audio version (especially the voices he mentions), listen to the NosillaCast podcast episode #868 which will be at podfeet.com/blog/2021/12/nc-868 when published on 23 December 2021.
Hey everybody Bodie here, I have a real quick tip for you. Recently, I had to go through a lot of documentation for my job. A lot of this is very technical information and it is not at all interesting. But despite my lack of interest in a lot of the subject matter, I still needed to know it, not even learn it, I needed to know it inside and out. The problem is, like many of you I have a full-time job, I have a family, I have other responsibilities, I have a house that’s you know aging and needs to be repaired, and things like that. And I didn’t have a lot of time to actually sit down and read. But despite my lack of time, I still needed to get through this material.
One day, while I was avoiding looking through all of these documents, I came across a program called descript. Now if you’re a podcaster, or you’re doing audiobooks, or anything with spoken word descript is a really cool program. Because you can either record your podcast or audiobook, whatever you’re doing into descript, or you can upload your audio and it gives you a transcription.
Now here’s the really cool part of descript is it gives you a transcription of everything that you said you can assign a speaker. So if there’s more than one person in your podcast, you can assign them. And then you get this transcript. As you delete things from the transcript, it deletes them from your audio file.
So it’s really cool that way, and also allows you to upload like your voice profile. So if you stumble, which I do a lot if you stumble, and you mispronounce a word, or maybe you say the wrong word, instead of like, if you wanted to say, Ford and you accidentally said fjord, or whatever you can, you can erase your mistake and type in Ford, and it uses your audio, your voice, to replace it with the correct word Ford, which is really cool. It’s magic, it doesn’t always sound great, but typically sounds really good.
One of the cool things about descript is they offer these canned voices, they call them speakers. So you can actually type something into descript. And this can voice will actually read it back to you. It’s pretty cool.
They have several different choices of speakers, they have some male, some female, you’re able to change the speaker’s inflections. So you can make them sound angry or happy or kind of like somber, or calm. It’s a really cool service. It’s not perfect, but I didn’t need it to be perfect, because I had a plan.
And that plan was to copy all of the documentation that I needed to go over in small segments into descript, and then use one of their canned voices to read it back to me. And then that way I could listen to it while I was doing something mundane at work or something mundane at home, which is a large part of my day at work and at home.
Or, if I was driving to work, it takes me 45 minutes to an hour to drive to work depending on traffic. That’s time that I could be listening to this documentation so that I can be learning it and then eventually knowing it. And a real quick spoiler alert here is it paid off because I really got to know the information at a level that I don’t think I would have gotten to if I was just reading it because I would have read it once or maybe twice, made some notes and then left it at that. So this was invaluable to me and what I was trying to accomplish.
Now I have a few samples of the voices for you. And I’m going to go ahead and play those. But before I play them, I want to let you know that the voices are very, very canned. Some of them sound better for different topics than other voices.
They give them names like Henry and Clara and Malcolm. I found Henry to be the one that worked best for me although sometimes depending on what was being read, I had to change the voice to make it sound better to my ears. But after a while, you forget the terrible canned voice and you really get into the information your brain adapts. So let’s listen to the first example.
The following is an excerpt from the Benefits in Burglar by Charles Neville Buck.
The agitated transit of Mr. Lewis cope well through the anteroom of the honorable Alexander Hamilton burrow created a certain stir
With all the lawless magnificence of a comet that runs amok through the heavens, he burst upon the somewhat promiscuous assemblage already seated there.
Real quick side jag, that book is in the public domain, so everything’s legal.
The first voice you heard was Malcolm, and use the one that let you know that this is an excerpt from whatever book, the second voice was Clara and you can kind of hear that she just kind of rushed through all of the text. You can make them slow down with periods and commas and things like that, but I just left it in so you get an idea. And the final voice was Henry’s, which, personally I thought sounded the best, but I’m a huge Henry fan. So I’m biased.
Like everything in life. There are positives and negatives and the script has some great positives. It makes it very easy to create an audio base out of nothing. So that’s really cool.
But it also has some downsides. And I’m not going to go through all of the downsides, but the biggest downside is that descript tends to be a resource hog. And as you add more and more content with your speakers, it really slows down the program. I
have a 2019 15 inch MacBook Pro, and descript can bring my computer down to a screeching halt sometimes. So this is one of those solutions where you’re going to have to use some elbow grease to accomplish your goal. And it’s not it’s not going to come without frustration. There is definite frustration in this in this solution.
However, when it works, it works really well. All right, I’m well over the time that I allotted myself to do this. So I’m going to go ahead and wrap it up here. Before I go I want to leave you with one more really fun descript example:
Kris Kringle here. I’ve been a fan of Allison’s and Steve since they were knee-high to a grasshopper. I just wanted to pop in and wish everyone a happy holiday season. What’s that Mrs. Claus? Yes, I’d love a delicious sleightini. By the way, my voice sounds different because I have a cold and not because I’m a computer-generated voice.
Did you notice there weren’t any ads in this show? That’s because this show is not ad-supported. It’s supported by you. If you learned something, or maybe you were just entertained, consider contributing to the Podfeet Podcasts. You can do that by going over to https://podfeet.com and looking for the big red button that says support the show.
Bodie asked me to to include the pricing from descript.com after his review:
descript has several pricing options. you can get 3 hours of transcription per month with full audio & video editing, unlimited screen recordings, and studio sound effect for free. I knew about descript but thought it was too expensive because I didn’t know they had a free option. If you have more advanced needs, you can get the Creator license for $12/month for 10 hours of transcription and watermark-free video export. There’s the Pro license for 30 hours/month along with lots of more advanced options and even an Enterprise license.