From experience, it seems no one actively tries to make their content inaccessible to their audience. Most people want to do the right thing, but they just don’t know how. After all, if the whole reason you’re making audio and video content is to reach and engage with large audiences, being accessible not only helps you to reach all the possible people your content can, but it also helps to show that you (or your brand, or your organisation) are putting your best foot forward when it comes to using technology to enhance the world around you.
In this article, I’ll step through how I made the captions for this video with AWS and no coding:
The Web Accessibility Initiative sum caption accessibility up perfectly - “Web accessibility is essential for people with disabilities and useful for all.” Video captions and subtitles are no exception to this, especially on LinkedIn where users are often browsing at work (or maybe even during meetings!), so they might not be viewing video content with headphones. Before AI/ML models got better and easier to interact with, this was a manual and time consuming task, but it’s much easier now.
Every time I post video content on my LinkedIn Profile I get asked how I make the captions, so I thought I’d put together a quick no-code tutorial. Let me know in the comments below if this works for you!
First, upload your video content to Amazon S3. If you’ve never used AWS before, don’t stress! Just think of S3 as Dropbox - the name S3 stands for ‘simple storage service’ which is all we need to think of it as in this tutorial.
Next, visit Amazon Transcribe in the AWS Management Console. To cut out some of the descriptions, I’ve recorded my screen making a little demo:
Essentially, you just want to find the menu on the right hand side and then select ‘Transcription jobs' and then ‘Create job’. From here, just follow the prompts. There’s lots of options to customise things or add features and you’ll notice I’ve used none of them in this video! If they’re something you’re interested in, or you’re going to be doing this at scale, I recommend taking a further look at the docs.
When the job is complete, the .srt file is ready in the S3 bucket I asked for, and then there’s a few ways to incorporate this into the video. Youtube and LinkedIn both let you upload .srt files when you share a video, and I really recommend this as a baseline.
If you’d like to get creative, you can also add the .srt file while you’re editing the video in Adobe Premiere Pro and most other video editing tools, which means you can change the font and placement of the text on top of your video. Here’s a quick example I made in Premiere Pro:
Now that you’ve generated the .srt file, you can use the text to upload a transcript (which helps with SEO as video content isn’t indexed in the same way as text via search engines) or you can use the srt output as the building blocks of a blog post.
Hopefully this quick tutorial on how to create video subtitles with Amazon Transcribe has been useful to you! If you try it out, make sure to tag me in the videos you create!
About the Author: Brooke Jamieson is the Head of Enablement - AI/ML and Data at Blackbook.ai, an Australian consulting firm specialising in AI, Automation, DataOps and Digital. Learn more about Blackbook.ai here and learn more about Brooke here.
Top comments (0)