In the evolving world of broadcast, staying ahead of the curve requires understanding and embracing cutting-edge technologies that are reshaping the industry.
Recently, I had the opportunity to sit down with Sepi Motamedi, who leads industry and product marketing for live media at NVIDIA, to dive into the trends that are driving this transformation. From the shift to software-defined infrastructure to the game-changing role of AI, Sepi shared her view on how these innovations are not only changing the way content is delivered but also unlocking new possibilities for broadcasters worldwide.
If you’re curious about what the future holds for the broadcast industry and how these advancements will impact your business, read on to see what Sepi had to say.
Q1. (Jeff) What are the trends that you’re seeing in the broadcast industry today and how do you see it shaping the future of broadcast?
(Sepi) The broadcast industry is currently shaped by two key trends: audience-led and technology-led developments. Audience-led trends emphasize delivering content where viewers are, across multiple platforms, with a focus on personalization. A prime example is the recent global sporting event, where content was delivered live, on demand, and in both long and short form across various platforms to meet the diverse preferences of audiences.
On the technology side, AI is increasingly driving innovation. While AI has been part of the industry for some time, it’s now ramping up in areas such as operations, monetization and accessibility, including powering real-time translation and transcription. The ongoing shift from SDI to IP and the transformation to software-defined platforms have opened the door to much easier integration of AI into media workflows. This evolution is expected to help redefine the future of broadcasting by enabling new capabilities for content creators and distributors, as well as enhancing the viewing experience for audiences.
A great example of this trend is the rise of auto-generated highlights. Just yesterday, I was talking to a friend about the recent global sporting event I mentioned earlier and asked if they had watched it. She mentioned that she hadn’t had the chance to watch the full events but instead caught up on the highlights that a broadcaster shared on a social platform. This is a perfect example of how AI can help media transform content generation and consumption, as highlights can now be generated by AI-driven solutions.
Q2. (Jeff) The industry has been moving itself towards software-defined broadcast. We’ve seen it at Dell as people moved away from bespoke hardware to software running on industry-standard IT servers. Looking at that transformation, what is your vision of software-defined broadcast, and what opportunities does it create for broadcasters?
(Sepi) That is a great question, and as so many of the people in the industry know, this transformation to software-defined broadcast certainly hasn’t happened overnight. This evolution has happened over many years with a lot of hard work and dedication by media companies, vendors and standards organizations around the world. What I am excited about is the recent technological advancements in this transition from SDI to IP. Initially, IP-based appliances increased flexibility and scalability, and now we are seeing true software-defined infrastructure with containerized applications running on commercial, off-the-shelf hardware accelerated with GPUs. All of this helps broadcasters streamline and consolidate workflows and infrastructure.
This evolution allows for greater flexibility, cost reductions and the ability to integrate AI into workflows more easily. Broadcasters can now build dynamic workflows, dynamically repurpose hardware and receive software updates independently of hardware, giving broadcasts the agility they desire without overpaying for agility they don’t need. Ultimately, this transition enables broadcasters to better meet varying demands without over-provisioning resources.
Q3. (Jeff) What do you see as the key components of software-defined platforms for broadcasting?
(Sepi) At the core of software-defined broadcast is the shift to accelerated, off-the-shelf hardware that supports any software. This requires effective management and orchestration of containerized applications, incorporating essential services like NMOS and PTP to ease the burden on application vendors. The infrastructure should be built on open APIs and industry standards like 2110, ensuring flexibility and future-proofing workflows as new technologies emerge.
NVIDIA is helping broadcasters in this transformation through the development of Holoscan for Media, a platform that embodies these key components. Running on NVIDIA-accelerated hardware, provided by partners like Dell, and using Red Hat’s OpenShift for orchestration, Holoscan supports a wide range of applications—from graphics and AI to mixing and transcoding. The platform’s open design empowers developers to create applications that the broadcast industry can spur on throughout this next evolution in the transition to software-defined broadcast.
Q4. (Jeff) Generative AI has clearly captured both the imagination and the headlines. This is a great opportunity to ask you about your thoughts about AI and the media industry. Specifically, thinking about the broadcast ecosystem, what should both broadcasters and broadcast vendors keep in mind when creating, adopting, or deploying new AI technologies?
(Sepi) From a developer’s perspective, it’s crucial to recognize the need for a clear roadmap to incorporate AI into solutions. Developing AI isn’t always straightforward, and many ideas get stuck in the prototype phase. To do it right, you need a platform that speeds up development, accelerated infrastructure, easy-to-use microservices, optimized models and enterprise-grade security to ensure stability. NVIDIA has built these tools and software, and is partnering with infrastructure platform providers like Dell to help media companies ease into and accelerate their development processes, helping them move from prototype to production as quickly as possible.
For media companies, the focus is on acquiring, engaging, and retaining audiences. With content from past and ongoing productions, as well as viewership data, media companies can use AI to create tailored content that resonates with their audiences. The question then becomes whether to build AI solutions in house or buy them. NVIDIA offers extensive developer resources for those looking to build, as well as a strong partner ecosystem for those who’d prefer to buy. Regardless of the path, accelerated infrastructure is key to deploying the latest AI solutions effectively and at scale.
“Insights from predictive AI can inform the accessibility of content, which may then be generated using video AI tools such as AI translation.” – Sepi Motamedi, Senior Product Marketing Manager, Live Media Solutions, NVIDIA
Q5. (Jeff) What are the use cases that you’re seeing and how are they benefiting broadcasters?
(Sepi) There are countless AI use cases, spanning areas like translation, transcription and enriched archives. We can categorize these AI applications into three main types: predictive AI, generative AI and video AI, with significant overlap between them. For example, insights from predictive AI can inform the accessibility of content, which may then be generated using video AI tools such as AI translation.
Specific use cases include audience analysis, trend prediction, and personalized content delivery, where generative and predictive AI intersect. We’re also seeing AI being used for budget analysis to determine if content spending aligns with audience engagement data. Auto highlight generation and automated video editing are other practical applications.
Additionally, AI is shaping fan experiences, such as using conversational avatars of pro athletes to engage with fans. Real-time translation and dubbing further enhance fan interactions. Ultimately, it’s about understanding what fans want and leveraging AI to deliver innovative, groundbreaking experiences.
Q6. (Jeff) Final question: How can broadcasters transition their workflows and get started?
(Sepi) To start, broadcasters can explore using software-defined solutions that run on advanced accelerated hardware. NVIDIA and Dell Technologies will be at the upcoming NAB Trade show, one of the world’s largest media and broadcast shows, in Las Vegas from April 5-9. At the Dell Technologies booth (SL4616), we will be able to share some of the solutions I’ve talked about here, have a conversation about your unique challenges, and share tips on how to optimize your individual AI roadmap going forward. You can even get in the NAB expo with a complimentary code: NS8349.