What are machine generated closed captions?

  Captions are something of a hidden gem. An underused resource that can help your video content reach a wider audience such as those who are hard of hearing or deaf.  What’s more, a startling number of people watch videos on mute. A study in the US found that up to 69% of users watch video content on mute in public places, along with 25% in private. Closed captions also give your SEO (Search Engine Optimisation) a boost and help get you up the Google pecking order. But you’re probably thinking “I don’t have time”. What’s the solution: Machine-generated closed captions.

But what are Closed Captions?

Closed captions go further than regular captions in that they show additional elements such as different speakers, music and other background sounds and descriptions. Regular speech is supported by non dialogue markers such as “laughs” or “cries”. Closed captions improve your site’s SEO and enhance the user experience.

Machine generated closed captions

Machine generated closed captions, also called automated closed captions (ACC) are created using automatic speech recognition (ASR) software. This cutting edge technology generates speech-to-text captions in real time. It can work both on and offline. While live ASR is used for television broadcasts, video calls and presentations, offline is useful for film captions and pre-recorded shows.

Human vs Machine

While machine generated captions are a cost effective and time saving option, they can compromise accuracy depending on the software used. ASR is poor at recognising technical jargon and proper nouns and is liable to misspell words. While this is not such a problem for everyday users, fully accurate captions are essential for those who are hard of hearing to understand the content. Machine services average at around 80% accuracy. Human captioning is far more precise, particularly around utterances and verbal fillers but comes at a higher cost.

Where are they used?

Machine generated closed captions are routinely used on live TV, for example sports coverage and news programmes. The space is also expanding with newer markets, such as video call platforms. Already Zoom, Google Meet and Teams all provide automatic captions during live meetings.

What does the law require?

In the UK, national broadcasters are required by the Equality Act 2010 and Section 303 of the Communications Act 2003 to make “provision for the deaf and visually impaired”. This is enforced by the regulator, Ofcom. In today’s internet dominated world, closed captions are essential for both inclusivity and accessibility. Whether automated or human based, there is a service to help you create user friendly content.

Categories...

Follow us...

Featured blogs...