Why AI has phenomenal content creation potential in M&E | Industry trends
In early 2021, IBC 365 took an in-depth look at the growing use of artificial intelligence (AI) and machine learning (ML) to streamline access servicesincluding the creation of subtitles and captions, and simplifying the transcription of interviews and other content for internal use.
The speed at which the trend was accelerating was concisely summed up by Anupama Anantharaman, Vice President of Product Management at Interra Systems: “We are seeing broadcasters using ML/ML-based platforms to automatically generate missing captions, check captions and audio alignments, ensure captions are correct, and maintain correct punctuation in captions.”
Eighteen months later, the use of these technologies to deliver a host of essential services is now embedded in many broadcasters – and it’s no wonder, when their accuracy (already often in excess of 90 %) is constantly improving. But now there are plenty of signs that AI is increasingly being applied effectively to other areas of content creation – from audio dubbing to image recognition and camera tracking.
Michel Bais is Managing Director of Mobile Viewpoint at Vislink Technologies, which acquired the former company in August 2021 and recently introduced a number of AI-based solutions.
“I see AI expanding into more and more areas of production, with many companies undertaking research to see how it can be used to make different tasks more efficient,” he says. “So, for example, we are already seeing AI being widely used for image recognition, as well as automatic camera tracking. It is clear that we are at the beginning of a very interesting chapter.
Sports Summaries and Player Recognition
From 4K to immersive sound, the sport remains an early and enthusiastic adopter of new technologies – and AI is no exception. Bais confirms that the continued expansion of sports services is fueling many innovations in this area, with AI-powered solutions helping broadcasters cover more events without unduly increasing their cost base.
“It’s not uncommon these days for rights deals to require you to broadcast a large number of matches. As a result, many production companies are looking to see how they can achieve this using lower-cost technologies that include an element of AI,” he says.
Specific examples include the automatic production of short content, summarizing goals and other “key points” of sporting events for publication and sharing online. Historically, says Bais, “many [these] the summaries were pretty basic. But with the AI, you have the ability to automatically recognize the goal, delete it, and also have all the reactions. This can be presented immediately to the publisher and made available soon after. So for this content, I see there’s a lot of potential for AI.
Vislink itself is gradually developing a considerable portfolio of AI solutions, including IQ Sports Producer, a live sports production and streaming solution that provides cost-effective, high-quality video coverage for professional and grassroots sports. The company has also developed an AI-based studio content production system called vPilot, which helps create professional-quality productions without a cameraman or directing team.
In addition to opening up new potential revenue streams by making it possible to cover a wider range of events, multi-camera AI production has the potential to provide “richer and more immersive” production than what could be obtained with a set of basic cameras. -at the top. With the AI studio production system, adds Bais, it is possible to add “engagement-building elements such as expert opinion and match analysis to programs. This helps content rights holders increase viewership and engagement levels, and deliver better returns on event coverage. »
This type of content is also suitable for distribution on multiple platforms, including social media, OTT, CDN, and live streaming for TV broadcast. In addition to being typically used for sports-related interviews, this type of solution can also be deployed in TV talk shows. “You can use a system like this to go live without a director or a cameraman, with the [image recognition technology] which allows the cameras to be positioned precisely, being careful not to take pictures of empty chairs, etc. So yes, we hear of people advocating the use of AI systems for these kinds of shows.
Slow start then exponential growth
Another pioneer of AI in media is media service provider Globecast. In fact, as far back as 2018, Shakunt Malhotra – Managing Director of Globecast Asia – identified 10 target applications ranging from quality control (QC) to sophisticated metadata creation, editing and compliance in a Blog for the Globecast website. Four years later, Malhotra remains more enthusiastic than ever about AI in broadcastwhile indicating that it will take time to realize its full potential.
“New perspectives are opening up every day,” says Malhotra. “We’ve been experimenting with using AI from QC to creating highlights. Recently we’ve been working on captions and virtual ads. However, I’ve yet to see a wider deployment of AI as it should be. Instead, we should look to another current industry trend for an idea of its likely trajectory.” I think it will be like the cloud, which had a slow start and then a exponential growth.”
Unsurprisingly, the use of AI to streamline the delivery of advertisements in different contexts is a particular target of industry activity. “Virtual advertising is another area where AI comes into its own,” Malhotra says, adding that “we’ve been working with our customers [to enable] replacement of the display boards in the stadium. With the help of AI, it can be more dynamic.
The technology is also used by online platforms to identify ad break structures, which can help ensure “a smooth transition to ads rather than the ‘choppy insertions’ that [can occur at present].”
Additionally, there are many applications in content distribution, such as predictive error detection. The use of AI in distribution networks to “identify patterns that can lead to incidents will help with good management of network bandwidth in CDNs as well as management of processing power in the cloud”. Another use case is the AI detection of color bars and clocks in live video recordings or legacy conversions – allowing those frames to be removed during transcoding before further OTT distribution.
As for Globecast’s own AI aspirations, Malhotra says the company’s strategy is currently focused on image/audio recognition and pattern recognition. “It is well recognized that the media supply chain will benefit. For example, speech synthesis and image recognition are already used for the management of multimedia assets. »
Implementing effective standards invariably accelerates the adoption of new technologies, and it remains to be seen whether there is a will to apply standards to AI in the way that appears to be about to happen in metadata managementNT. Meanwhile, it’s a safe bet that we’ll see continued refinement of AI-powered technologies so they can deliver the kind of rock-solid reliability that broadcast engineers seek.
Humans aren’t 100% accurate and neither is AI, Malhotra says, “but it has to get closer [than it is at present]. For this to happen, more data needs to be added to ML so that there is a wide variety of use cases. [that can feed back into AI implementations]. Once we have this critical information, it will be possible to get closer to that idea of perfection that broadcast ops managers love to see.
AI is seeing rapid adoption in broadcast environments as it offers substantial cost and efficiency benefits. Learn more about Broadcast AI.