All podcasts should include transcripts for the people who need them or those who prefer to read than listen, but a lot of podcasters are operating on a very low budget and can’t afford to add additional costs. Fortunately, there is a solution, coming from a company called Headliner with a product called Eddy, and I will give you a step by step description of how to do it.
Note: I am not affiliated with the company in any way. I just want to help more podcasters offer transcripts for their listeners.
First, go to https://eddy.headliner.app and create a free account. When you first log in, you’ll see a page that looks like this.
For this guide, I’m going to create a transcript of an episode of Stephen West’s Philosophize This! podcast. It’s a great podcast for learning about a lot of different topics on philosophy, and I just listened to an episode featuring Phillip Goff about Pan-psychism. I went to my favorite podcast app, PocketCasts, and downloaded the audio file as you can see here.
Next, I dropped it into the Eddy Drag & Drop or Browse field, and the page changes to a waiting screen while the file is uploaded.
When the upload finishes, the following screen appears.
If English is not your language of choice, you can choose another, then press CONTINUE WITH THIS LANGUAGE and the AI transcription will begin.
In order to keep the product free to use, Headliner has come up with a clever revenue model, in which advertisers can display ads which play short clips of other podcasts while your episode is being transcribed. You could always mute these and go to another tab, but you might find a new podcast you like if you go ahead and listen!
After a few minutes, depending on the length of the audio file, the AI transcription is complete when the following page appears.
Clicking on the GO TO PROJECT button will reveal this screen.
Name my project creates a title for the transcript derived from the content, Add chapters to my project creates subheadings at appropriate intervals, and Remove filler words does what it says. I haven’t really tried the last toggle, or adding my podcast yet. For this guide I’m just going with the defaults, so the next step is to press CONTINUE WITH THESE SETTINGS.
Here we can see the title Eddy made from the content it transcribed, and you can use this or choose from a variety of options.
You might notice that the speaker name is not the host of the podcast, but a generic “Speaker A.” If you click on the button, you can edit either a single occurrence of all occurrences of this label.
This podcast episode has a second speaker as well, and you can do the same to change it.
If you just want to read the transcript right away, you can press the play button on the audio control at the bottom and the words being heard will be highlighted while it plays which looks like this.
If you’re more interested in just creating a readable transcript, click on the EXPORT button at the top of the page.
Next, you’ll see the Export your Project modal, where you can choose to export the audio or a text transcript. The second is what we’re after, so click the Transcript button.
The next screen allows you to select a file type; TXT, PDF, DOC, VTT, or SRT. I recommend TXT or DOC, because you’re probably going to want to clean up a few basic mistakes. The AI is really good, but humans still have a role in this beyond just clicking buttons. It will often make grammatical errors, misspell proper nouns, and will not have any paragraph breaks, resulting in a “wall of text” output that isn’t very reader friendly. I’m going to export a TXT file and go with the default settings, the click EXPORT YOUR TRANSCRIPT.
Sure enough, we can see some errors in the resulting text file.
Our host would probably prefer his last name to be capitalized and for the podcast to be named correctly. Re-opening the file in MS Word, I can easily fix those kinds of mistakes, add paragraph breaks where necessary, and then style it appropriately and with my preferred title matching that of the episode. The final step is to create a PDF that is optimized for accessibility. Choose the “Best for electronic distribution and accessibility” option, as it is fine for printing and also will play better for anyone who might need to read it with a screen reader.
Here’s our final PDF. I’ve also chosen to make the URLs into real links for anyone who might be accessing the final electronically and wants to jump directly to the sites mentioned.
Now you know how to make a quick PDF transcript using Headliner’s EDDY. Headliner has a lot of other features you can look into in your own time, including generating promo clips and help with promoting your podcast. Go to https://headliner.app to learn more.