I Learned About Audio Media Accessibility (And More!) From Improving Tech Podcast Transcriptions

Hello Fellow CodeNewbies 👋,

During Hacktoberfest 2021, I was browsing and looking for an open source to contribute to. I only had a little time at that time. So, I limited my search to accessibility and documentation issues. And that was when I found an accessibility issue. It was on Michaela Greiler's Software Engineering Unlocked Podcast repository.

The issue was to improve the transcript of a podcast episode. We must read or listen to the episode, fix the typos, and add missing words. My first thought was, "Wow, this is interesting."
Accessibilities issues that I often found have things to do with visual-related problems. So, audio media accessibility was new to me. And to my surprise, I learned a lot from improving podcast transcriptions.

Audio Media Accessibility

The Web Content Accessibility Guidelines (WCAG) has a section designated for audio and video media.

I quote here from the guidelines:

... a person who is Deaf can't hear the audio, so you need to provide important audio information in another form.
— Web Content Accessibility Guidelines (WCAG).

Why don't you try this? Read an auto-generated podcast's transcription or YouTube's closed caption. And don't listen to the audio.
Do you notice that sometimes the transcription is out of context or funny?

I'm not an English native speaker. I often read transcriptions or turn on the subtitles when I watch a video. It helps me understand things better.
Some people enjoy reading the transcription rather than listening to the podcast. They will listen to it when they can't understand the transcription.
But let's imagine we're deaf or hard of hearing and entirely depend on text to absorb information. How can we get that information out of the audio media if the transcriptions or the subtitles are hard to digest?

Auto-generated transcriptions are great as the base to provide accessibility. It's also cheaper than hiring someone to transcribe.
But they still need to be improved by humans. Most of the time, you can see typos that make the transcription even harder to understand.

Let me give you an example from one of the episodes I improved on the Software Engineering Unlocked Podcast. It's the episode with Harshit Chitalia, the co-founder and CTO of Tromzo.

And I don't have to think about logging because there's an open source. System that can do it at already for me, a package that I can use. Right. And then we have like, luck for Jay and the vulnerabilities and so on. Right.

When I read that, I asked myself, "Okay. I know nothing about security and DevOps. But what is this luck for Jay? Although it sounds off, is it one of their terms or packages?"

So I reread the whole paragraph and listened to the audio. I still couldn't get it. Then I googled everything that sounded like luck for jay. I added some keywords like login, DevOps, etc., until I found out that the speaker was talking about Log4j.

Then the transcription became:

And I don't have to think about logging because there's an open source system that can do it at already for me, a package that I can use, right? And then, we have Log4j, and the vulnerabilities, and so on, right?

Now we can understand what the speaker was sharing with us.

Here is another example I shamelessly took from the episode I'm in on Virtual Coffee Podcast:

So, I mean, I need like three to four hours to arrive. 600 to 800 words. And you imagine so.

I was in this episode. But I had to laugh a bit because it sounded funny. Did I say it that way 🙈?

Then I listened to the episode, and here is what I said:

So, I mean, I need like three to four hours to write 600 to 800 words. Can you imagine? So ...

After listening and giving proper punctuation, the transcript made more sense.

From these examples, we can now see how important it is to improve podcast transcriptions and video subtitles. Having someone to enhance transcription is essential. They can add missing words and search for words that a speech-to-text app can't pick up. They can make all the efforts to make the transcription readable and understandable.

People who depend on texts have the right to access and get the correct information.

Search Engine Optimization (SEO)

I was improving a Virtual Coffee Podcast episode with Matt McInnis. There was a moment when Dan, the host, talked about his Phillips hue lights and how he connected them to the phone. Then Matt said:

I hope all the -- our conversation in this podcast has an index cuz it'll go to these Google searches. I mean, this is gonna be all Philips hue light bulbs.

From there, I learned that improved transcriptions also mean better SEO.

Here is another example from the episode with Harshit Chitalia:

So one of the real use cases for using TRAs or for our customers was trying to find out where all log four J exist in their environment.

And here is the improvement of the above's transcription:

So, one of the real use cases for using Tromzo for our customers was trying to find out where all Log4J exist in their environment.

In the above case, the speech-to-text app picked up TRAs or instead of the company name, Tromzo. The company's typo is all over the place with different words. There is Zo, proso, Trusso, and many more. Suppose we let these typos remain there. In that case, Tromzo wouldn't get picked up by the search engine. And people won't get the information the speaker wanted to share here.

Improving transcriptions gives a good chance for the podcast — and the episode in particular — to crawl and be indexed by Google. The better the transcription, the bigger the chance for it to appear on the top search result on Google or other search engines.

Tech World In General

I'm learning about web development and JavaScript in particular. I have little knowledge about tech outside of what I'm learning.
When improving a transcription, I have to read, listen and re-check the transcription. Through this process, I gain more information and knowledge about the tech industry. Reading the transcription and listening to the podcast opened my eyes. And learning about people's journeys in tech also became a motivation for me to move forward.

Final Words

Web accessibility is broad. And one of the things we can do is to make audio and video media accessible to everyone.
So, friends, if you're a podcaster or a video content creator, start providing good transcription and subtitles. It will make your content accessible. That way, the information you want to share can reach everyone. Besides, doing so would help with the SEO of your content 😊.

Thank you for reading!
Last, you can find me on Twitter and Mastodon. Let's connect! 😊

📸 Cover image by Jonathan Velasquez on Unsplash