In an era where content is king, the integration of VIDA Content Os platform by BBC and Getty Images marks a significant leap forward in the audiovisual industry, especially in the realm of archive management and content monetization. This collaboration, underpinned by advanced AI capabilities, is set to redefine how media assets are accessed, managed, and leveraged, offering an unprecedented level of efficiency and innovation. The upcoming interviews in TM Broadcast International delve deep into the technical intricacies and visionary approaches behind this monumental project, shedding light on the challenges overcome and the breakthroughs achieved.

TM Broadcast’s readers can anticipate a comprehensive analysis of the seamless integration of cloud technologies, AI-driven transcription services, and user-centric platform design that collectively enhance the accessibility and utility of vast media archives. The firsthand accounts from key figures behind the initiative, including Simon Roue of VIDA Content Os, Chris Hulse from BBC, and Paul Davis, Vice President at Getty Images, EMEA, will provide valuable insights into the strategic decisions and technological advancements that have propelled this development.

This feature article aims not only to inform but also to inspire industry professionals about the potential of AI in transforming content management practices. The discussions will explore how the VIDA Content Os platform’s implementation facilitates a more intuitive search, curation, and distribution process, thus setting a new standard for content platforms in the digital age. Through this analytical lens, we unveil how this pioneering project stands as a testament to the power of collaboration and innovation in harnessing the full potential of AI in the audiovisual sector.

BBC

Recently, we’ve known that VIDA developed a MAM system to manage and make accessible 57,000 BBC documents to the public. These files were only available searching by hand, and now AI capabilities are supporting and managing these documents in a efficient way, for the public’s benefit.

First, TM Broadcast International speaks with Chris Hulse, Head of Motion Gallery at BBC Studios.

Chris Hulse has been the Head of BBC Studios Motion Gallery since 2014 when they partnered with Getty Images to represent the BBC’s TV and radio archive for commercial clips sales. He has over 35 years’ experience in the media industry, having been an Archive Researcher prior to leading teams, digitisation programmes and technical projects within the BBC. His varied roles have given him practical experience of working with film, videotape, audio, music, stills and file-based media. Chris is very proud to be facilitating the re-use of the BBC’s exceptional programming globally whilst providing valuable support to their public service broadcasting role.

Regarding the recently implemented VIDA Content OS, can you provide more details on the collaboration between Getty Images and BBC Studios in launching this new platform for accessing BBC archive video content?

BBC Motion Gallery and Getty Images combine our expertise in content and sales to commercially represent the BBC’s TV and radio archive across the globe. We pride ourselves on our excellent customer service and openness to feedback. When asked to make our content more readily accessible we took this on board, investigating the specific requirements of our clients. The resulting project was time-consuming but well worth the investment to deliver a highly regarded product that is having a positive impact on our businesses and our clients.

How does this platform enhance the customer experience in terms of searching, purchasing, and downloading BBC archive video content?

Our previous library platform only gave our registered clients the opportunity to view our content. If they wished to select clips for licencing they needed to be sent a secure link for each programme, watch it again and clip up sequences which they wished to licence. With VIDA they are able to research, browse and clip up content 24/7. They only need to contact the Sales team to obtain additional research, or to discuss master clips and licencing, which saves them both time and money.

The platform allows customers to securely search the entire digitized library. Can you tell us about the security measures implemented to protect the BBC archive content?

The VIDA platform contains all the content ingested by Motion Gallery and is only accessible to our registered users. The registration is limited to our commercial clients and managed internally. The site is not open to web crawlers or search engines. The high-resolution master content is not directly accessible to our clients. Their clips are manually released only after licencing.

How the curated collections enhance the creative process for producers and programme makers?

The collections are not aimed to be exhaustive but offer programme makers a short cut to content curated by our expert researchers which covers commissioning trends and upcoming events. This may provide valuable insights for current projects or inspire new ones.

We have read that post-clearance high-resolution masters are available immediately.

How does this benefit projects with fast turnaround times, and what impact does it have on production timelines?

When clients select clips from our preview files the platform generates high resolution versions at the same time. This means that there are delays in master fulfilment. Once our client has signed the licence the clips are able to be downloaded straight into their edit.

How do the latest Speech to Text tools integrated into the platform facilitate transcript search for all assets, and how does this contribute to uncovering hidden content from the BBC Motion Gallery’s vast archive?

The platform runs speech to text transcription at the point of transcode. This provides us and our clients with valuable additional search options for data that is not available via the BBC Archive’s metadata set.

Can you provide insights into the significance of making iconic BBC content, such as The Office, Absolutely Fabulous, BBC Sports Personality of the Year, and Planet Earth, available for download for the first time ever?

Our aim is to make the BBC’s amazing archive more visible to the global production market. Highlighting iconic programmes and brands brings footfall to the site where clients are then exposed to our fantastic editorial content which they may previously been unaware of.

How does this new platform contribute to the licensing and distribution of BBC programmes globally?

Motion Gallery’s business is licencing clips from the BBC’s vast TV and radio archive. The VIDA platform actively supports this by making the BBCs programmes more visible to our global audience who may not be aware of the breadth of the BBC’s output. This may then lead to sales leads for the BBC Studios programme distribution team.

How does VIDA Content OS address the growing demands in the archive footage licensing industry, providing an integrated and efficient supply chain for servicing the catalogue, particularly with the use of cloud and AI technology?

Behind the scenes we have moved away from local spinning disc and tape robotics to a managed cloud solution. This not only delivers financial savings and greater sustainability but a much more robust and effective service to our clients. Wherever they are in the world they will have the same user experience.

With the BBC being over 100 years old it has a vast archive. This has been managed via many different systems and teams over the years which impacted the level and accessibility of metadata. Utilising AI tools to deliver reliable transcripts for our content at the point of transcode give programme makers a much richer source to draw upon. Identifying contents such as personalities and speeches which were not previously identified.

How does the partnership between Getty Images and BBC Studios aim to assist programme makers around the world, and in what ways does the VIDA platform play a crucial role in achieving this objective?

Our partnership aims to make the BBC’s quality programming and editorial content readily accessible to programme makers globally. We leverage Getty Image’s worldwide sales team and provide clients with a trusted archive platform that is available for them 24/7, helping them to locate specific shots or ideas for commissioning pitches.

Is BBC planning new developments related to its contents and its management?

Motion Gallery is focused on enhancing our offering to clients. Our aim for the year ahead is to develop our VIDA platform to deliver more content and further enhance the user experience.

Getty Images

Part of this collaboration enterprise, Getty Images provides its experience working with VIDA and BBC throughout the VIDA Content OS’ implementation process with Paul Davis, Vice President at Getty Images, EMEA; Davis answered TM Broadcast questions about the latest advance in the field of the archive management.

How did the company realize that it required a new MAM solution?

Our goal is to help Producers get their productions made – VIDA allows us to collaborate creatively with our clients at development stage to build programme ideas and seamlessly gain access to over 57,000 digitised programmes.

The key realization was driven by the need to improve our overall customer experience and efficiency as well as enhance access to the incredible BBC Motion Gallery library. The previous MAM solution only provided registered users with a view-only platform experience and created barriers between our customers and our content. Our goal was to breakdown those barriers and allow customers the ability to research and partially self-serve in real time 24/7, which massively benefits our customers dealing with fast turnaround times.

How long did it take to create and implement the new MAM solution?

Conversations around the launch of the new VIDA platform began in 2018 but it took a significant amount of time to bring this to life, with BBC Motion Gallery, Getty Images and VDMS developers working tirelessly together to build the best solution possible for our clients.

What kind of archives (AV, only audio, only video) are managed by this new MAM platform?

The platform currently supports the BBC Motion Galleries AV content offering, but we are working towards expanding to also include BBC Radio Archive on the platform.

What kind of issues has it resolved?

The VIDA platform allows customers to research content from over 57,000 digitized assets from the BBC library and go directly to master selection on any of these featured items- this digitized collection is also growing every day. Content is being added both proactively and as part of continued offline research access available to our customers. Any BBC programme that has Library sales rights, and has not previously been digitized, is now added to the VIDA platform library as soon as the new preview is completed.

We have also been able to integrate the VIDA Content OS Platform with Getty Images internal processing systems to allow for a more streamlined content ordering and delivery process – this has reduced turnaround times and administrative load. This integration also allows for the real time release of reproduction rights and copyright information with master content delivery, which was not possible on older platform solutions.

Does the VIDA Content OS platform offer any new feature that makes easier managing and monetizing the content?

The VIDA Content OS Platform allows customers to view and research watchable content freely, view meta-data, pull transcripts, store, and share collections of content, download previews and order master clips against their projects, all of this leads directly to more content consumption and licensing, whilst making the collection easier to support, reducing turnaround times and administrative load.

The platform also provides curated collections – this allows our clients to view content that is matched to what is selling in the market to allow for deeper creative collaboration in the development stages. Another very exciting feature for industry professionals is the ability to utilise the new speech to text transcriptions which will allow programme makers to uncover content that may have previously been difficult to unearth.

This new MAM solution, could it be used to manage all Getty content?

This MAM solution offers our clients with a bespoke improved and enhanced solution to the BBC Motion Gallery offline collection. Getty Images represent more than 551,000 contributors, of which 80,000 are exclusive to Getty Images. We will continue to explore all solutions to meet the needs of any customer—no matter their size—around the globe.

However, Getty Images’ website already offers a growing library of over 551 million visual assets (video + stills) that delivers unmatched depth, breadth, and quality – an extensive and comprehensive archive (historic & full range) with over 135 million images dating back to the beginning of photography and our video collection contains over 26 million video clips and growing. The BBC Studios online collection is also available online with over 200,000 licence-ready clips.

What were the main difficulties or challenges when implementing the new VIDA Content OS solution?

Our number one priority was putting the customer at the heart of everything we do. The constant consideration was ensuring this solution resulted in the most positive outcome for all parties engaging with the platform. Balancing the needs of the BBC MG archive, Getty Images internal processing systems and ensuring that this was a legitimate upgrade to the previous customer experience.

Until now, what MAM solution did Getty use for its data base?

Previously the BBC library offered clients a view-only experience. This platform was completely independent to the client ordering process. This new platform offers clients a full end-to-end journey from allowing clients to research and partially serf-serve.

What role did the royalties or reproduction rights / copyright when implementing the VIDA MAM platform?

We had to consider how reproduction rights information and copyright info was provided to customers using all content on the platform. There is a dedicated team of experts working at both BBC Motion Gallery and Getty Images to ensure that all content released onto the platform is diligently checked and approved before release. The VIDA platform allows for master content to be released to customers with all clearance information attached in real time and ready to be ingested into an edit.

How did the VIDA platform integrate with your former CRM? Are both systems interrelated now?

Before September 2023, VIDA did not integrate with any system at Getty Images. Our CRM is now fully integrated with VIDA, which allows customers to preview, clip up, order, track requests and download master content directly via VIDA platform. This has revolutionised the customer experience, providing self-serve abilities, improving time to access content and enhanced accessibility to our content.

Recently, Getty developed together with Nvidia a Generative AI tool for content ideation and creation, how does it work?

We are excited to offer our customers the Generative AI by Getty Images tool which pairs the company’s best‑in‑class creative content with the latest AI technology for a commercially safe generative AI tool. Generative AI by Getty Images is trained on the state‑of‑the‑art Edify model architecture, which is part of NVIDIA Picasso, a foundry for generative AI models for visual design. The tool is trained solely from Getty Images’ vast creative library, including exclusive premium content, with full indemnification for commercial use.

Customers simply type in the image prompt, use the offered filters around content type, aspect ratio and colours/mood and click ‘generate’. Four generations are received off every prompt and from there customers can go on to refine what the generator created and see what pre-shot imagery exists around the prompt in case that gets them to the desired image faster.

Has it learnt from your archive or database?

The tool was trained on Getty Images’ vast creative library, including exclusive premium content, driven by our global creative insights research and-curated by our internal experts who elevate the best authentic and representative imagery.

What is the main goal to achieve with this new tool offering?

To offer our customers a high-quality tool which won’t put them at legal risk. Not all other AI generator services and tools are automatically safe to use for commercial purposes. Uniquely, we can say ours is as it is exclusively trained off our premium creative content, so you won’t run into legal issues around using images which contain likenesses of real people, trademarks, logos, works of art or architecture, or other elements protected by third-party intellectual property rights that you do not have the right to use.  We also believe that a better-quality training set gives you a better-quality output.

Which kind of developments can we expect from Getty Images in a near future?

We’ve just announced advanced inpainting and outpainting features now available via API. Developers can seamlessly integrate the new APIs with creative applications to add people and objects to images, replace specific elements and expand images in a wide range of aspect ratios. These features will soon launch on iStock.com and Gettyimages.com.

VIDA

TM Broadcast spoke with Symon Roue, VIDA’s CEO, to know how the main challenges affronted and which breakthroughs comes with this project’s development.

Which were the requirements for this project and how do you adapt the solution VIDA content to this project?

Access to the BBC’s extensive library is not direct due to the nature of the broadcaster being a public service entity. To facilitate library searches, Getty Images, through its commercial relationship with the BBC, is contractually permitted to sell content from the BBC library. The complexity arises with the library being only partially digitized and the challenge of surfacing content from the undigitized portion.

As a researcher, initial access through Getty might reveal a selection of clips from their public website. To delve deeper, an account in VIDA is created by Getty Images, which unlocks further levels of access. A significant endeavor was the migration of 57,000 assets into VIDA, necessitating transcoding and the creation of proxies for each asset. Furthermore, to enhance searchability within the platform, we undertook the transcription of over two million minutes of content.

Integration with Getty Images’ CRM, specifically Salesforce, was also critical. This allows users to search the VIDA library, select clips, and add them to their basket, which in turn generates an opportunity in Salesforce. However, not all clips are immediately licensable due to potential third-party content or rights clearance issues, such as those involving the Royal Family. Once the rights checks are complete and commercial negotiations finalized, users can utilize watermarked proxies for their edits, and upon clearance, download the high-resolution asset from VIDA.

The platform also caters to instances where content is not yet digitized in VIDA but exists within the BBC’s library, including various departments like BBC archives, news, or sports. These departments can upload files into VIDA, enabling them to be surfaced for licensing. This sophisticated system offers an easy-to-use environment for researchers to surface previously inaccessible content and for Getty to fulfill its commercial obligations, ultimately generating revenue for BBC studios.

This journey to commercialize the BBC’s rich archive and provide access to over 5,000 researchers through VIDA has been successful, though it was a complex process not to be underestimated.

It’s a lot of work. How long did it take to develop the platform?

It was about six months. We’ve been working with the BBC for a little while prior to that, but the integration with the rights management system, the Getty Images development team also had to be involved because they’re opening their CRM system for us to be able to integrate with. It wasn’t without its challenges, but we overcame all of them.

How does the platform work from the point of view of the user?

VIDA guides users into a frictionless environment via Passwordless entry, simplifying access with a secure link sent to their email, or through various single sign-on services. Upon entry, users are greeted with a curated homepage by BBC and Getty, highlighting thematic collections that span diverse categories, including natural history.

These collections are a cornerstone feature of VIDA, enabling both our hosts and users to assemble and personalize content compilations that can be shared with collaborators—all within the platform’s secure confines. This curation is complemented by a robust search engine that penetrates beyond surface-level data, utilizing transcription-based metadata to unearth previously inaccessible assets. This capability allows footage researchers to conduct granular searches with keywords extracted via sophisticated speech-to-text tools, revealing valuable clips for their projects.

Our integration extends to the shopping experience within VIDA. Users can select clips, tailor them to their project’s scope, and seamlessly add them to a basket. This initiates a process within Getty Images’ CRM system, ensuring that rights and commercial considerations are validated. Once the requisite checks and negotiations are completed, users are notified that high-resolution assets are ready for download, streamlining their creative process.

It’s important to note that VIDA’s scope extends to content not yet digitized within BBC’s archives. We’ve established integrations with various BBC departments, ensuring that any requested content can be promptly uploaded and made available on our platform. This sophisticated yet user-friendly system underscores our commitment to providing a state-of-the-art content repository and distribution service, benefiting both the BBC’s extensive archives and our commercial partners at Getty Images.

Which were the main challenges for this development?

Certainly, reflecting on the main challenges we encountered while developing VIDA, it becomes clear that our cloud-native approach was a decisive factor in mitigating potential obstacles. The sheer volume of content, exceeding a petabyte, could have presented a significant hurdle if we were constrained by traditional on-premise methods. Instead, leveraging the expansive compute power of AWS and utilizing Dolby’s advanced transcoding engine allowed us to efficiently process and transcode a vast archive of media.

Synchronization among stakeholders was paramount. Ensuring alignment across the teams at VIDA, BBC, and Getty Images, while maintaining business continuity, was a complex dance deftly managed by our adept project managers and customer success team. The adoption phase post-launch brought its own set of user-centric challenges, from login issues to email discrepancies – the typical teething problems of any large-scale digital transformation.

Yet, it’s the people aspect, the change management, that truly tests the mettle of technology deployment. Witnessing users move past the initial adjustment to embrace the full potential of what VIDA offers, and receiving their enthusiastic feedback, has been particularly gratifying. It underscores the success of the platform not just in technical terms, but in its ability to revolutionize user experience and content accessibility.

We suppose that, at least at the beginning, the work had to involve BBC and Getty, but which departments were involved? Do you set a special team for this work?

The development of VIDA was a collaborative effort, drawing on the strengths of multiple departments across Getty, BBC, and our VIDA team. At Getty, the synergy between the commercial, development, and marketing teams was pivotal. Their efforts in user education through webinars and training were crucial for a smooth transition to VIDA.

At the BBC, the rights management and commercial teams faced a paradigm shift in their daily operations, managing the risks associated with technology change and ensuring customer satisfaction.

Within the VIDA team, the customer success, development, project management, and marketing departments played integral roles. Communicating internal changes and fostering a collaborative environment were key responsibilities, especially for our new team members like Philippe, who joined us to strengthen our internal communication.

The development was a testament to teamwork, with each department contributing to the collective goal and ensuring a seamless introduction of VIDA to our users.

Which are the main differences between managing video / audio files and pictures? How did you overcome these differences and integrate all documents across the same system?

VIDA is proficient in handling a diverse array of documents and images, yet this specific collaboration was focused on the BBC’s rich trove of audiovisual content, which includes video, audio, and associated metadata. Despite VIDA’s capacity to manage both still and moving images, our primary objective here was to facilitate the licensing of the BBC’s audiovisual assets. We successfully integrated content across various resolutions, from standard definition to 4K, and even 5K for certain unedited raw footage, particularly from natural history archives. The versatility of VIDA ensured a seamless user experience regardless of file size, which is integral when dealing with such a wide spectrum of content.

What compression protocols did the team work with?

All of the source assets are stored predominantly in the ProRes format. That’s all the mezzanine assets. Then the proxy files, which is the viewable, streamable content, that is all using H.264. There are various bitrate ladders, so depending upon the internet connection that you are working with, the system will dynamically decide which bit rate is optimal to be served to use to give you the best searching and screening experience. By keeping it relatively simplified and not going wide on the different formats, you are able to build a much more easy-to-integrate-with kind of system.

This MAM platform uses AI features and the latest speech-to-text technology, could you tell us a little more about this state-of-the-art tech and what capabilities / possibilities offer?

VIDA harnesses cutting-edge AI models from industry leaders like AWS and OpenAI, specifically the Whisper technology, to offer unparalleled speech-to-text capabilities. These models have advanced remarkably over the past few years, now adeptly handling multiple languages and regional dialects with greater sophistication. Our platform leverages these tools to produce high-fidelity transcripts, providing options to balance speed and accuracy depending on project requirements. We can even isolate speech from background noise, thanks to the sophisticated pre-processing these models facilitate. By continuously integrating the latest AI advancements, VIDA ensures that content is future-proofed, empowering users to reprocess assets with new models as they evolve, extracting even more value over time.

What can you tell us about working with the cloud? What are your providers for storage?

In the realm of cloud-based operations, AWS stands as our principal storage solution, providing a robust and reliable foundation. To ensure maximum resilience and disaster recovery, we also engage with additional providers such as Wasabi, which allows us to maintain triple redundancy for our clips across a diverse array of cloud services. This multi-tiered approach assures long-term security for our stored content. Additionally, we utilize Amazon CloudFront for its global reach, ensuring users anywhere can experience swift and smooth playback. Our infrastructure also includes Aspera’s high-speed transfer capabilities, ensuring rapid content delivery that can fully utilize available bandwidth. The elasticity of our system is another hallmark, capable of scaling seamlessly to accommodate any number of users instantly, thanks to AWS Lambda functions. This sophisticated infrastructure represents the pinnacle of modern cloud technology, offering a virtually limitless operational scale.

Getty has a different platform for its clients, with a larger data base, Are you involved any how in its platform’s creation and / or managing?

Getty Images operates its own platform, designed for immediate licensing and e-commerce capabilities, which provides clients with readily available content. While I am not directly involved in the management or creation of Getty’s platform, our collaboration becomes pivotal when clients require deeper, more specialized content. That’s where VIDA comes into play, granting access to those who seek beyond the surface, such as documentary filmmakers needing unique and not widely accessible footage.

What future developments are you preparing?

In our roadmap for 2024 and 2025, we are focusing on deepening AI integration to enhance content context analysis within shows. This means moving beyond keyword searches to more intuitive queries, like identifying images without explicit mentions in the script. AI will also aid in summarizing show content, providing concise synopses that streamline search efforts. Additionally, we’re developing an advanced player to facilitate navigation through enriched metadata markers. Expect to see more dynamic, user-centric search functionalities, akin to e-commerce experiences, bringing efficiency to content discovery and tracking. Our goal is to refine user experience, ensuring time spent on VIDA is both productive and enjoyable.