How Captions Enrich Video – Engagement, Reach and Accessibility

By Calum Davidson, Strategic Product Manager @ Red Bee Media

It was recently announced that Red Bee is providing captions for the DPP’s online events, and in the press release the DPP expressed the importance of providing accessible content to its community.

Enriching content in this and other ways was in fact also the main topic when I took part in the DPP innovation week last month, where I gave a three-minute pitch in the Content Enrichment slot – novel ways to get more from your video.

I was watching the other pitches like a hawk to see what might happen should anyone go over time – no-one did, disappointingly. Professionals all round.

Having helped organize captioning provision many times in the past, I’ve now had the opportunity to experience an event like this from the other end. After years of worrying about producing accurate captions, I’ve now had my own words transformed to text on a screen. I apologize to those trying to deal with my ridiculous word rate – I blame the clock sharing my screen, with its faintly alarming three-minute countdown.

Our pitch focused on highlighting the benefits of adding captions to broadcast or streaming content.

We know for example that captions:

• Increase engagement – with people being more likely to watch and watch for longer
  • 69% of people report watching without sound in public places
  • 80% say captioned video means they’re more likely to stick with until the end
    Source: Verizon/Publicis
• Increase reach – making content easier to categorize, search and easier.

Of course, these two points are important to content owners and how they make the best use of their assets, but we can’t get away from the fact that the fundamental job of captions is to allow those watching media to understand and enjoy it. The increased comprehension that captions create is enriching the experience for the viewer.

Given the nature of the topic that day, it felt a bit like we’d snuck this end-user angle in under everyone’s nose. In the same way, nobody has yet commented on my using a cobbled-together list of misheard lyrics/mondegreens as a visual reference for “textual comprehension”.

Gaining metadata through captions is a benefit we’ve talked up for a while now, but what is new is the fact that content providers now have so many choices in how they obtain those captions.

There is now a multitude of options, including human live captioning through re-speakers or stenographers, as well as offline captions transcribed and timed by a human. With the introduction of automatic speech recognition (ASR) technology, captions can be created automatically and aligned for sync, with a human editor/QC stage. And finally, now that technology has caught up to the point where it’s viable, high quality automated live captions. Our Arc system uses human captioners’ experience to curate and filter an ASR engine, resulting in broadcast-level-accuracy captions.

Still, with all these available tools and methods to create captions, an Ofcom report on the first half of 2020 found that “On the platforms where[…]access services were supported, 66.3% of programming hours were subtitled”.

I was surprised at this number, especially considering how accessibility enriches the viewing experiences for audiences generally, and how captioning benefits content owners themselves. If you then consider the wide variety of captioning options, automatic or otherwise, it’s even more surprising we’re not closer to 100 % on all platforms.

As an aside, it’s interesting to see that Ofcom report also has a note saying: “We are aware that the national lockdown from March 2020 had an adverse effect on the production of access services.”

2020 was indeed a year like no other, but it’s been noted by several appreciative clients that, throughout the year, we had superb availability across Access Services. It’s partly thanks to our excellent tools that allow us to maintain that level of service. Of course, far more important, it’s down to the hard work of people at Red Bee to keep our clients’ content accessible – those who take the “enriching” of content through captions (or, for that matter, Signing, or Audio Description) very seriously indeed.

Finally, as a reminder to myself if I speak at the next DPP event, for the benefit of those trying to caption me if no-one else, I’ll try not to be so rushed by the alarming three-minute clock.