AI-Powered Live Subti­tling: Acces­si­bility for Modern Media

The demand for captioning has grown exponen­tially with the rise of global media consumption and stricter acces­si­bility regula­tions, such as the European Acces­si­bility Act (EAA), which will come into force in 2025. While this is of great benefit to around 10% of Europeans, the rules for providing acces­si­bility vary from service to service.

For public broad­casters, the regulatory require­ments are very high compared to private broad­casters, according to an EU-wide study. This study also quotes an estimate of the cost of providing acces­si­bility: “The provision of broad­casting acces­si­bility services results in signif­icant additional costs for broad­casters, ranging from 1% to 30% of production costs”.

Over the past few years, we have done several live subti­tling projects with private broad­casters, where the regulatory require­ments may be less demanding, but where there has been a lot of progress over the past few years.

Tradi­tional manual subti­tling has its place, but it remains labour-intensive and costly, especially when it comes to subti­tling live broad­casts. There is still a long way to go before we reach the point of being able to deliver closed captions (with sound effects ) via AI, but a 24/7 live news channel will not be able to offer manual live subti­tling, and this is where we come in:

AI live captioning is an efficient, scalable solution that can dramat­i­cally improve the avail­ability of captions for live and on-demand content. With our dedicated archi­tecture, we are able to provide 24/7 AI subti­tling, combining the transcription with powerful dictio­naries and deliv­ering it via linear teletext.

The benefits of the aiconix AI subti­tling via Teletext

  • Improved acces­si­bility:

    The upcoming EAA regula­tions will make it mandatory to provide subtitles for audio­visual content. Aiconix’s AI captioning technology helps broad­casters and streaming services meet these require­ments, ensuring that content is acces­sible to people with hearing impair­ments, as well as a wider global audience. Learn more about acces­si­bility in media.

  • Speed and scala­bility:

    Unlike manual captioning, which takes time to complete, AI-powered live captioning can generate captions in real time, day and night. This is partic­u­larly beneficial for live broad­casts such as news, sports and events, where immediate acces­si­bility is critical.

  • Cost-effectiveness:

    The cost of manual subti­tling can be prohib­itive for many media providers, especially for live regional content. However, AI signif­i­cantly reduces these costs, enabling imple­men­tation at a fraction of the cost.

  • Broadcast grade:

    Designed for broadcast workflows, our system not only provides you with an archi­tecture that enables 24/7 transcription, but also with automated recovery options in the event of a failure.

  • Integrated workflow:

    Subtitles are delivered directly to teletext via the inserter, providing viewers with the same proven experience, while AI transcription is seamlessly integrated behind the scenes.

Deep Live Hub: Lever­aging AI for Live Subti­tling via Teletext

The DeepVA platform and its Deep Live Hub enable real-time transcription of spoken content into text, providing the basis for instant subtitle gener­ation. The platform’s AI models are trained to recognize speech in multiple languages, making it highly effective for live trans­lation and transcription across multiple media formats.

Techni­cally, DeepVA’s AI models process the live audio stream and generate subtitles that can be delivered via HTTP Live Streaming (HLS) and via CC, BurnedIn and AMT (our JSON based format). For integration into tradi­tional broadcast systems, Aiconix works with SDI insertion technologies from partners like Peter Albrecht Elektronik. The Albrecht Inserter facil­i­tates the embedding of these AI-generated subtitles into an SDI signal using teletext standards.

This system works seamlessly to integrate subtitles into both HD and SD signals, allowing broad­casters to serve audiences on a variety of platforms. Using this workflow, we are able to subtitle one of Germany’s leading 24/7 live news programmes, even at 2am, whilst keeping the costs low.

DeepVA and Deep Live Hub: Lever­aging AI for live subti­tling

The future of audio­visual media lies in the ability to equally reach approx­i­mately 10% of European audiences with disabil­ities. Offering your programs to all possible viewers isn’t just a matter of compliance; it’s about fostering inclu­sivity and sustain­ability by ensuring no group is excluded. Acces­si­bility also enhances staff motivation, knowing their work is acces­sible to all. And there are small victories too: Imagine being on the subway, where everyone can enjoy content without needing headphones and annoying those around them, thanks to AI-generated subtitles.

But subti­tling is just the start. With DeepVA as Composite AI platform, we can simplify dialogue for broader compre­hension, or, through start-ups we bring to our platform, we offer automated dubbing in multiple languages or automat­i­cally improved dialogues for more acces­si­bility. This makes content more acces­sible while also unlocking new markets by supporting multi­lingual subtitles, catering to both global audiences and regional minorities.

As AI services rapidly improve in quality and function­ality, tools like Deep Live Hub not only solve today’s subti­tling challenges but also future-proof your content for acces­si­bility laws like the European Acces­si­bility Act, ensuring you’re ready for the demands of tomorrow.

AI-powered live captioning is not only cost-effective but also essential for meeting upcoming acces­si­bility regula­tions like the European Acces­si­bility Act, ensuring compliance while enhancing audience reach.

Share

Email
LinkedIn
Facebook
Twitter
Search

Table of Contents

latest AI news

Subscribe to our newsletter

Don’t worry, we reserve our newsletter for important news, so we only send a few updates once in a while. No spam!