Video Marketing

Human Moderated or Auto-Generated Video Closed Captioning – What’s the Better Choice?

With the ever-growing rise of video content in today’s digital world, video-closed captioning has become more important than ever. 

Brands and content creators have had to adapt to the emergence of including captions in videos. By doing so, they’ve had to choose between auto-generated captions – created using automated speech recognition (ASR) and human-moderated captions – created and reviewed by real people. 

Social media platforms like TikTok and YouTube offer technology to auto-generate captions for videos, making the process quick and easy. But, going the auto-generated route can come with limitations that human content moderation can solve. 

So, how do you know which type of captioning to use? Before we get into the pros and cons, it’s crucial to recognize why we need captions for video, to begin with. 

Why Do We Need Closed Captions in Video Content?

Accessibility

Closed captions make video content accessible to people with hearing impairments. 

In Canada, 60% of the population aged 19-79 have a hearing health problem. And according to a study by Facebook, 41% of videos are meaningless without sound or captions. 

Consider how much of your audience you miss out on because they can’t access your content. Including captions is a simple way to make your videos accessible and inclusive. 

In Canada, 60% of the population aged 19-79 ahve a hearing health problem. And according to a study by Facebook, 41% of videos are meaningless without sound or captions.

Video SEO 

Search is evolving and creating new opportunities for businesses that use video. So if you’re looking to supercharge your search results, video SEO is the right strategy. 

One of the first things you should do is add transcripts and captions to your videos. Search engines crawl and index video captions, meaning you can rank for a variety of key phrases mentioned in your video beyond what you’ve optimized for in the title and description. 

Closed captions are also proven to increase engagement. For example, a Facebook study found that captions increased video views by 12% compared to non-captioned videos. This is because captions keep viewers engaged even if they can’t hear the audio.

Content Moderation

Captions are an easy way to moderate video content – especially user-generated content (UGC). We know UGC is in popular demand by brands, but it can come with risks. For example, suppose your customers say something that goes against your brand compliance. Captions are one way for content moderators to review and flag videos before posting them online. 

Including video closed captioning adds value to content. But there are two ways to create them. 

So, What’s Better – Human Moderated or Auto-Generated Captions for Video? 

Here’s our comparison breakdown based on accuracy, cost and turnaround time. 

Accuracy 

The most significant difference between auto-generated and human-moderated captions is accuracy. 

It’s not hard to believe that real people deliver higher accuracy than ASR technology that automatically captions videos. However, while some ASR captioning services claim 96-99% accuracy, it’s imperative to consider how the accuracy is defined and how long it takes to achieve. 

When it comes to generating accurate captions, human content moderators are at an advantage. 

This is especially true when the audio quality is poor. Real people can understand language and make sense of conversations even if the audio is muffled. Auto-generated captions tend to fall short of accuracy when the audio isn’t 100% clear. This leads to inaccurate captioning, requiring human content moderation to fix. 

What if the speaker has an accent? Computers that auto-generate captions have more difficulty understanding varying accents, speech patterns and pronunciation than real humans do. This is because computers need to be programmed with specific accents to interpret the content accurately, whereas humans can do so much more naturally. 

Cost

The main reason why many brands turn to auto-generated captions is cost. 

Using ASR technology is much more cost-effective, mainly because you don’t have to pay computers for their time. 

However – the cheaper costs come with lower quality and lower accuracy. Besides, when brands opt for the auto-generated option, they usually need a real person to moderate content and make changes where necessary, which is an added cost. 

Turnaround Time 

There are two kinds of turnaround time to consider in video closed captioning – engagement and display time.

For engagement time, chances are auto-generated captions will be faster as there is no need to wait for a person to be available. However, auto-generated captions need a content moderator to review for accuracy.

For display time, we’re referring to live captions that appear on the video screen. But, again, auto-generated captions need content moderation by a real person to check for inaccuracies and make changes where needed. 

While auto-generated captions may seem faster to produce, they always require a real person to review, which can delay turnaround times. 

What’s the Better Choice?

Both auto-generated and human-moderated captions come with their share of pros and cons. However, neither one should be substituted for the other. In most cases, brands can achieve optimal closed captioning results using a combination of auto-generated and human-moderated captions.

StoryTap Does Content Moderation for You

At StoryTap, we make the video-closed captioning process easy! Our in-house content moderation team will review your videos for accuracy and make the appropriate changes on your behalf.

Using our patented video technology, you can create authentic video content that is accessible, search optimized and content moderated all in one. 

Take advantage of what StoryTap has to offer today! Speak to our team and learn how we can help your brand with video. 

Want more content like this, straight to your inbox?

Subscribe to our monthly email roundup of helpful marketing resources.