Providing transcripts is one of the primary way to increase your podcast’s accessibility. There are a few ways to create transcripts, and a few ways to display them.
Creating transcripts
Doing it yourself
- Pros: Cheap, greater accuracy, potentially faster turnaround.
- Cons: Time consuming
When I started my podcast, I was doing my own transcription. The transcripts were, out of the box, the most accurate transcripts I ever got - Obviously I understood what I said and knew the topic we were talking about. It didn’t cost me anything but my time. And I knew exactly when I was getting the transcript back.
However, it was quite time consuming, particularly as I wasn’t familiar with making transcripts.
There is software available, often for free or at low cost, that allows you to run the sound file and pause it as you type. I won’t recommend any in particular here as these things change fast. It also often comes down to personal preference. Do a search for “transcription software” and you’ll have many options. Beware! Some of these applications are using speech-to-text engines rather than allowing you to do manual transcription.
Automated transcription services
- Pros: Relatively affordable, quick.
- Cons: Poor accuracy.
More and more services use speech-to-text engines to create transcripts. These services are generally quick. It takes a matter of minutes to get transcripts for hour-long show. However, the accuracy is generally not good. I tested several of these services and I personally chose not to use them.
They can be a good option if you want to provide transcript but you are strapped for cash. Some would say: “Better a transcript accurate to 80% than no transcript at all”.
You can use automated transcription services and then manually correct the errors. This can be a good balance between cost, speed, and accuracy.
I have used Temi with some success in the past1. They currently charge US$0.10 per minute of sound file. There are other services.
Human transcription services
- Pros: Relatively good accuracy
- Cons: Potentially expensive, potentially long turnaround
Human transcription services are the most expensive way to go. They are also the most accurate. They require the least effort from you. While quite accurate, they are rarely 100% accurate.
They can also be quite expensive. I’ve seen rates as high as US$5 per minute of sound file. The best balance I’ve found has been from Rev2. They charge US$1/minute and are generally quite accurate.
If you need transcripts for languages other than English, the rates increase very quickly.
Buyer beware. Always do a test run of 1 episode before committing to multiple episodes.
Displaying transcripts
There are a few ways to display transcripts. The best way is directly in the page, right after your show notes, in plain text or HTML.
If you provide transcripts, always say so in the intro of your podcast episode. This allows people who access your podcast through apps rather than your site to know transcripts are available.
Plain text or HTML
- Pros: Content available immediately, increases SEO
- Cons: May detract from the look, may not be implementable on your podcast platform
This is, by far, the best way to provide a transcript. The information is right there for everyone to see and skim if they are looking for specific information.
As it is part of your podcast’s episode page, it is also available to search engines. This will ultimately increase your reach.
An example of inline transcripts for the A11y Rules Podcast: https://a11yrules.com/podcast/gaad-2018-special/
Another example of inline transcript for The Podcast Host: https://www.thepodcasthost.com/podcraft-podcast/interviews/accessibility/
Downloadable file or linked transcript
- Pros: Easier to upload.
- Cons: Harder to access for podcast followers, no SEO advantage, often poorly accessible
If you are receiving a document from a transcription service, it is easy to simply upload that document and provide a link to it.
While it is easier for you, it is more difficult to get to the transcript for your audience. They have to download a document. It splits their attention between the transcript and the show’s page.
Downloadable documents are also rarely fully accessible, particularly if you are providing PDF files.
An example of a linked transcript for The Big Web Show: http://5by5.tv/bigwebshow/176
Synchronized
- Pros: Can be followed along with audio
- Cons: Complex implementation, possible expensive
Synchronized transcripts are interesting because they allow someone to both follow the audio and the text at the same time. Of course, they can also just read through the transcript and then listen to the bits they find more interesting.
The implementation of synchronized transcript is more complex and will not be discussed here. Just know it’s an option!
An example of synchronized transcript for the Hansel Minutes on Greta.com: https://gretta.com/hanselminutes/
-
I am not affiliated with Temi in any way.↩
-
I am not affiliated with Rev in any way.↩
As an expert in the realm of podcasting and accessibility, I have a profound understanding of the various methods employed to enhance a podcast's inclusivity, with a keen focus on providing accurate transcripts. My experience extends beyond theoretical knowledge, as I have actively engaged in podcast creation, transcription, and the implementation of accessible practices.
The article you've presented underscores the pivotal role of transcripts in bolstering a podcast's accessibility, and it skillfully navigates through different approaches to create and display them. Let's delve into the concepts introduced in the article:
1. Creating Transcripts:
a. Doing it Yourself:
- Pros: Cost-effective, higher accuracy, potential for faster turnaround.
- Cons: Time-consuming, requires effort.
b. Automated Transcription Services:
- Pros: Relatively affordable, quick.
- Cons: Poor accuracy.
- Note: Some services utilize speech-to-text engines, sacrificing accuracy for speed.
c. Human Transcription Services:
- Pros: Relatively good accuracy, less effort.
- Cons: Potentially expensive, potentially long turnaround.
- Recommendation: Rev, at US$1/minute, strikes a balance between cost and accuracy.
2. Displaying Transcripts:
a. Plain Text or HTML:
- Pros: Immediate content availability, SEO benefits.
- Cons: May affect aesthetics, platform implementation challenges.
- Examples: , .
b. Downloadable File or Linked Transcript:
- Pros: Easier to upload.
- Cons: Harder for followers to access, no SEO advantage.
- Example: .
c. Synchronized:
- Pros: Can be followed along with audio.
- Cons: Complex implementation, potential expense.
- Example: .
3. Additional Considerations:
- Always mention the availability of transcripts in the podcast episode's introduction.
- Beware of automated transcription services' lower accuracy, but they can be a cost-effective starting point.
- Human transcription services offer higher accuracy but may incur higher costs.
By providing this comprehensive breakdown, I aim to empower podcasters with the knowledge to make informed decisions regarding transcript creation and display, ultimately contributing to a more inclusive podcasting landscape.
