While the 2021 NAB Show has been moved to October, the spring season will still feature a cavalcade of new product releases and groundbreaking news coming out of the broadcast technology sector. In an effort to keep the video-production community informed, SVG is hosting a series of SportsTech On Demand video interviews throughout April, May, and June with executives from the industry’s top technology vendors.

Russell Wise, Digital Nirvana, SVP, Sales and Marketing, addresses how sports operations are challenged like never before to deliver engaging video content to their viewers in the tightest-possible timeframes and the way Digital Nirvana’s portfolio of AI- and ML-based solutions are helping sports organizations around the world to streamline critical processes – from generating transcriptions to delivering compliant closed captions and beyond.

A study from Valuate Reports found that the global captioning and subtitling market to reach US$ 466 million by 2026 from US$ 277.3 million in 2019, at a CAGR of 7.7. The report further states that the United States of America and Europe would be the largest markets for captioning and subtitling solutions. Artificial Intelligence (AI) is playing an essential role in the growth of automatic captioning solutions.

However, before we proceed, what is AI? In simple terms, AI is a simulation of human intelligence in machines programmed to think like humans and mimic their actions.

AI is vital for closed captioning because, in its absence, the whole process is highly time-consuming. Organizations that have leveraged AI-based captioning solutions and workflows can vouch how the manual process that took 10+ hours is now performed within a couple of hours with accurate captioning and transcription. For media and broadcast organizations, which produce tons of content, AI-based closed captioning solutions help them to reduce the captioning turnaround time, enhance efficiency, and keep up with the FCC regulations, among other benefits.

AI State of Affairs in Captioning:

We live in a world dominated by Siri, Alexa, chatbots, and personalized search results, all of which rely on AI technology. However, it is essential to know it’s just not AI; it is AI, ML, and ASR technologies. Let’s understand each one in brief:

• Artificial Intelligence (AI): Refers to human intelligence replicated by machines.
• Machine Learning (ML): Refers to the workflows allowing machines to learn from previous experience.
• Automated Speech Recognition (ASR): It converts speech-to-text.

Today, we are showcasing how Digital Nirvana has successfully applied AI to media workflows to increase the speed of productivity and accuracy. In this case, our client is one of the leading providers of short-form content and entertainment news, and they had a specific set of business challenges. Our client had the requirement to take 20 hours of video footage and develop a 20-minute show out of it with a turnaround time of 2 hours. They also had to generate accurate transcription to enable their editors to quickly locate the content of interest and edit it into a show and once the show is developed, generate closed captions in English and Spanish, again all in the turnaround time of 2 hours.

This is where AI technology comes into play; Digital Nirvana’s captioning solution Trance leverages AI and ML technologies to deliver automatic captioning. Trance can process hundreds of hours of video footage, using speech-to-text technology to create accurate transcripts that allow editors to go in and make the edits. Trance comes with an automatic transcription generator that delivers results in 30 different languages allowing the broadcasters to provide captions in more than 100 languages for worldwide content creation and distribution.

Download – Trance whitepaper can help you understand the benefits of AI for captioning

How Trance facilitates automatic transcription, translation, and captioning:

• Media ingestion through various sources, from the production asset management system or from a cloud location or directly uploaded to a particular portal such as Trance.
• Trance then generates speech-to-text and presents it in a word editor form that could be easily accessed and processed. This allows the users to search content easily and fix the error if any.
• Our system is enabled in generating transcripts, a preliminary process to create captions. Users can use commonly-used parameters preset in the system after the transcripts are generated so that the system automatically converts this transcript into a form like this, where you can define whether it’s two-line captions or three-line captions.
• As mentioned in the above use case where the client needed to generate different language captions, all they must do is click Add Language to access a dual-pane window. The user now has access to the source video, the source language captions, and an automatically translated version of the other language.
• The system can auto-generate transcripts in more than 30 languages, and you can translate captions to over 100 additional languages. Process these captions into your automation system, and voila you are done!

This entire process that earlier took some 12-15 hours is now reduced to a task accomplished in under 30 minutes, and this is made possible by AI. By leveraging Digital Nirvana’s AI-enabled speech-to-text (STT) and translation engines, broadcasters and content creators can enhance their content creation and distribution with 99% accuracy. You can contact us here, and we will be happy to take you through a demo and showcase how our AI solution can transform the way you create content.