TEDx Talk – High Quality Captions and ROI

Above is an YouTube recording of Sveta’s talk about high quality captions and ROI at TEDxFultonStreet in New York City in summer of 2018.

The video is captioned, but a full text transcript is also posted below with visual description for those who cannot see the video or prefer to skim text. The transcript is complementary, not a replacement to video captions.

Want to learn more about captioning access? Want to hire Sveta to consult your organization on how to improve experience of captioning access or to speak at your event? Contact us. Looking forward to working with you!

Full text transcript with visual description

(Visual description: Video starts with a slide saying: “TEDxFultonStreet x=independently organized TED event.” Then transitions into a slide with an image of “Yes or No” with arrows showing they are interchanged. Then transitions to a recording of Sveta signing on a stage with a slide behind her saying: “Svetlana Kouznetsova, Accessibility Expert.”)

Sveta voiced by a sign language interpreter: 

“If you create video and audio content or host events, there are two reasons to care about providing high quality, same language captioning and transcription. The first is obvious, that there are millions of people like myself, who get accessibility through captions. But another reason is that you will reach a much larger audience and increase your return on investment.”

(Visual description: A slide with an image of 8 people below of mixed gender and race looking above the space that has text appearing and saying: “80% of captioning users are NOT deaf – source: UK Office of Communications (OfCom)”)

Sveta via the interpreter:

“80% of people use captions who are not deaf. Captions help foreign language learners, people in gyms, or bars, or sound sensitive environments, and assist when the speaker has a strong accent or when content is complicated and difficult to understand.”

(Visual description: A slide with an image of hands holding a smartphone in front of a laptop. Text appears saying: “Videos WITH captions: 40% – viewership increase (Source: PLY Media)” then transitions to text: “Videos WITH captions: 90% – watched to completion (Source: PLY Media)” Then transition to a slide with an image of hands holding a tablet and text saying: “85 of Facebook videos are played on MUTE”)

Sveta via the interpreter:

“Captions increase the viewership by 40%. 90% of videos with captions are watched to completion. Digiday states that 85% of Facebook videos are viewed in mute mode, so captioning has become a necessity.

Growing up, I didn’t have captioning access until I was in eighth grade. That’s when my dad brought home a box called a captioning decoder. We were amazed by this magic box that made captions show up on almost all the TV channels. (Visual description: A slide appears showing a captioning decoder.) Now captioning support is included in all TV models and video players, and there’s no need for external decoders anymore. Captions are now not limited even to TV and videos, they can also be provided in real-time for breaking news, webinars, events, classes, work meetings, and so on.

Once I was given access to captioned material, my life completely changed and new worlds opened up for me. But I’m not alone.

(Visual description: A slide with an image of a back of an young girl sporting hearing aids and holding an iPad. Text appears and says: “USA: over 48 million (Source: Archives of Internal Medicine, 2011)” and then “World: over 466 million (Source: World Health Organization – WHO).”)

Sveta via the interpreter:

“There are over 48 million deaf and hard of hearing people in United States. And over 466 million worldwide. The number is increasing rapidly as well, especially because the number one cause of hearing loss now is constant exposure to loud noises.

Captions also helped me master English as my third language. Even my parents, who have perfect hearing, love captions because English is not their native language.

Quality is critical when you’re considering captions. While there are now many speech technologies that can translate automatically spoken words into captions, they’re often not accurate. Many deaf people call them “craptions”.

(Audience laughs)

Under ideal conditions, the accuracy rate is usually somewhere between 80 and 90%. And it gets even worse with complex content, or accents, background noise, when speakers are speaking too fast or overlap each other. And bad captions are not better than nothing. Garbled captions are hard to read. And it’s just like listening to garbled audio. Caption errors make it hard to understand or retain any information. 

Poor captions affect your SEO, search engine optimization, and decrease your CPM, your clicks per mille, for videos and your ROI, return on investment. Google is deaf like me and doesn’t like videos with auto-captioning, so it just does not index those videos. So Google penalizes producers for turning on the auto-captions instead of actually adding proper captioning themselves.

Even if a machine could translate every spoken word to captions accurately, it still cannot follow certain quality guidelines. For example, you may see text exactly as it’s heard, for example this: lets eat grandma.”

(Visual description: A slide with an image of text “lets eat grandma” on the left and of an old lady on the right gasping.)

(Audience laughs)

Sveta via the interpreter:

“Now, does that mean that we should go ahead and consume grandma or that we should dine with her? It’s not clear, right? Now that’s much better.”

(Visual description: A slide with an image of text “Let’s eat, grandma!” on the left and of an old lady on the right holding a tray of cookies, smiling, and holding a thumb.)

(Audience laughs)

Sveta via the interpreter:

“So proper grammar and punctuation are critical for optimal experience in reading captions. Just like speech intonations are.

Quality guidelines are long and complex, so I’m just sharing a few basic tips with you today. Transcripts and captions need to be verbatim, they need to follow proper grammatical rules, include non-speech elements, use upper and lower case letters, have good font size and color contrast, and translate profanity word for word, assuming it’s not bleeped out in the audio. Captions and transcripts need to be completed before you publish or share videos and podcasts. So everyone can enjoy them at the same time.

And if you advertise in advance that events will have real-time captions, you will get a larger audience.

With advances in technologies, it is much easier, faster, and cheaper to create captions now more than ever before. For example, YouTube offers a great free tool to create and edit captions. You can save and download your captions as an SRT caption file and import it onto Vimeo,

Facebook, or other video platforms that offer captioning support.

Here’s a short example.”

(Visual description: A quick video demo showing how to type in captions in YouTube.)

Sveta via the interpreter:

“So you can see how easy it is.

Yeah, it’s really easy. And you can clean up the auto captions if you like, or you can create them from scratch. Just please do not turn on auto captions and just leave them as is. And please add video captions on all platforms that you use and couple those captions with transcripts for videos because those are useful for people who cannot see the video or that wanna skim the information.

If your organization can do so, consider hiring an experienced specialist who is familiar with quality standards and guidelines. They can produce your captions and transcripts faster and make them look more professional and easy to read. The costs for transcripts and captions vary from a dollar all the way to  $10 or $15 per minute. And the rates really depend on experience and the skills  of the vendor, the audio quality, the content complexity, among other factors. Low cost may result in bad quality. Some vendors may use auto-captioning themselves or unskilled labor that will result in low caption quality and end up wasting your money. Always review the selected vendor’s prior work to assess the quality. If you have budget limitations, there are some creative ways to find money like asking sponsors to cover expenses.

I work with media producers, corporations, educational institutions, business owners, event organizers, and I help make their audio, video, and events accessible at a reasonable cost. There’s some guidance and information along with my personal story that can be found in my book.

If nobody asks for speech to text access, that does not mean that there’s no demand. It needs to be provided for all types of aural information even without a prior request. High quality, same language captions and transcripts are good for business and a necessity for millions of people.

Let’s think outside the ears!”

(Visual description: An image of an young girl holding her ear with text saying: “Let’s think outside the ears!”)

(Audience laughs)

Sveta via the interpreter:

“Let’s caption everything!”

(Visual description: A profile of an young girl with text to the right depicting the sentence)

Sveta via the interpreter:

“Get your message to more people!

Thank you.”

(Audience applause)

(Upbeat music)