Subscribe to NewscastStudio’s newsletter for the latest in broadcast design and engineering delivered to your inbox.
In part two of our Industry Insights roundtable on media asset management, our experts weigh in on topics including changing approaches to MAM, changes to metadata and search, content repackaging and broadcast workflows.
How has continued WFH changed your approach or thinking on MAM technology?
“The requirement for remote working hasn’t changed our approach as much as it has reinforced our view that leveraging cloud-based workflows provides the most efficient access to content and applications for people that may be located anywhere. Media management, therefore, must be able to move content through a media supply chain that integrates automated cloud-processing with manual workorders for operators who may be still working from home,” said Geoff Stedman, CMO for SDVI.
“At Dalet, it has been our vision to enable remote workflows for years. We have been actively implementing and delivering solutions that enable content creators and distributors to complete even the most complex tasks from home. With WFH now being part of everyday life and flexible work arrangements being an expectation, we are designing our solutions with mobility, agility and flexibility as key drivers. All our solutions, whether for news production or media workflows, are cloud-enabled, and hence our clients can adjust their infrastructure set up based on their needs,” said Mathieu Zarouk, director of product strategy, Media Workflows at Dalet.
“WFH has challenged production and post-production in how users access media efficiently at every stage of the workflow. WFH has accelerated the evolution and adoption of cloud-based infrastructures for content management as organizations look for ways to enable collaboration between locations and to reduce their reliance on on-site data centers. Having the flexibility to adapt to these changing times has never been more important,” noted Savva Mueller, director of business development for Telestream.
“Media production workflows are not reverting back to traditional models. Over the past 18 months, MAM technology providers have been forced to prioritize roadmap items, pushing collaboration, workflow, remote visibility and discovery features to the top. These new enhancements are a direct response to the new normal where teams are not heading back to shared facilities but they still need to collaborate with one another as if they were sitting in the same cubicle or edit bay,” said Rick Young, head of global products at LTN Global.
“The pandemic has amplified the problem of how the media and entertainment industry grapples with managing media assets remotely. Demand for content is booming exponentially, with audiences confined to homes under lockdown, putting further stress and strain on the ecosystem. Meanwhile, the show must go on with people working from home without access to traditional hardware or infrastructure and audiences demanding entertainment. We have been extremely busy over the past 18 months, helping broadcasters, studios, and content owners meet these challenges by migrating their workflows and libraries to our cloud-native media supply chain platform,” said Bill Admans, COO for Ateliere.
“Our approach to MAM technology aligns with our overall strategy for the Grass Valley Media Universe and our Agile Media Processing Platform (AMPP), which was in place before the pandemic. While people continue to work from home, it has simply supercharged our customers’ appetite for tools and workflows which support it,” suggested Camilla Powell, a product manager for Grass Valley.
“Remote access has been part of our software DNA for many years. By providing web and transfer-based access to media from craft editors, our platforms are ready for the work from home environment. This is a requirement we envision will be key for many years to come and have committed to provide increased features in this model such as access to live ingesting feeds directly from the web interface to make it easier to review and log baseband signals,” said Alan Dabul, product manager of MAM & media I/O for Ross Video.
“It’s absolutely more important than ever. As everyone abruptly shifted to a WFH model, it became immediately obvious to companies in all industries that collaborative software and the internet were going to be the glue keeping the world moving and the distributed workforce connected. Communication and organization—two things a good MAM can facilitate well—are the keys to a highly effective team, and those two things become immensely more important when team members are spread across multiple locations,” said Eric Newbauer, VP and CTO for Studio Network Solutions.
How will new technology evolve metadata, search, and content repackaging?
“The continued improvements in machine learning and other AI tools to analyze media content and extract more information about the content creates a wealth of new metadata that aids in content search, modification, and repackaging. Media supply chain platforms can use this metadata, associated with specific timecode, to create time-based metadata, which helps operators quickly find specific points of interest in a piece of content (ad breaks, black, slates, objectionable content, etc.). The result is that more content can be processed more quickly, for more delivery endpoints,” Stedman explained.
“Early adopters are already using machine learning, natural language processing, and other AI capabilities in terms of generating relevant metadata, structuring, as well as in distribution. We’ll see large scale adoption of machine-generated metadata and structured organization from content creation through archival – and everything in between – as these will become essential for finding content internally by organizations as well as serving the customer with relevant content. For time-sensitive content such as news and sports, we’ll see content metadata being generated in real-time for raw footage to aid in search and retrieval of accurate content faster, thereby reducing time to market,” said Russell Vijayan, head of business development for Digital Nirvana.
“Due to increasingly data-driven media production, the need for content-based automated metadata generation is also increasing. As AI-based analyzers are becoming commonplace in media workflows, we have moved from a position where we had relatively small amounts of metadata that we trusted, to a situation where we have huge amounts of metadata but have varying degrees of confidence about the accuracy of that data. Managing these confidence thresholds is key to the usefulness of the metadata,” said Karsten Schragmann, head of product management for Vidispine.
“I expect increased automation and better standardization. AI will play an increasingly important role in content auto-discovery and recommendations. Gradually, technology will also deliver better insights to drive decision making, especially for content commissioning,” Zarouk predicted.
“Advancements in AI and machine learning are providing access to details of media that were previously only accessible via manual entry. With features such as face recognition, both historical and new media can be enriched through automated workflows, thereby providing a level of searchable access that has never been available. The newly available data will facilitate a quicker production environment and enable media owners to sell and distribute easily,” Mueller said.
“As hosted infrastructure costs — bandwidth, process and storage — continue to decrease, the ability to perform certain functions will grow. For example, rich AI-driven metadata capture, which is processor intensive, will become more practical and affordable. Deeper metadata on ingest (and delivery) will streamline processes for creation and also drive monetization specifically tied to the ability to put exactly the right content and supporting advertising in front of the right consumer at the right time,” said Young.
“At Ateliere, we utilize technologies like AWS Rekognition to scan and log every aspect of content libraries, making it easy for producers, editors, and reporters to quickly find scenes, a task that was previously cost and time prohibitive. We have developed proprietary AI that analyzes content libraries to identify duplicates and eliminates storage redundancy, helping customers save high costs. Our technology also tags content segments as individual components and creates virtual versions without rendering whole files, significantly reducing the storage footprint. We scan audio files to create subtitles and captions in a matter of moments and accurately match a human translator. The bottom line is that artificial intelligence and machine can help content owners save millions and profit significantly from new monetization opportunities,” Admans answered.
“There is a real breadth of offerings for AI services on the market to augment metadata, search, and content repackaging. One of our products allows our users to select services and providers for themselves, rather than being tied to a single supplier selected by us. The key to taking advantage of this evolution will be well thought out user experiences and workflows to translate the technology into actual efficiencies,” said Powell.
“AI and remote access will help accelerate the creation of metadata for many workflows including search and content repackaging. By using AI services to create speech-to-text transcriptions along with translations, content can be localized at a highly reduced cost and effort of work. By providing remote access, customers can utilize talent in remote locations to amplify the number of users working on their platforms from geo dispersed locations,” Dabul said.
Where are we headed? To a cloud-native supply chain model?
“There are already excellent examples of existing cloud-native supply chains, such as A+E Networks, Discovery, ViacomCBS, WarnerMedia, and more, who are embracing this journey to the cloud. These visionaries recognize that moving to the cloud not only helps them become more efficient, but it gives them newfound agility to pursue new opportunities much faster. And with the right platform, they get true insight into how much resource and cost are associated with every piece of content,” Stedman said.
“Only a year ago, one of the biggest decisions for anyone looking to implement, replace or upgrade part or all of their media supply chain solution was where to put it – on-premise, in a private cloud, in the ‘public’ cloud, or some hybrid of two or all three of those. Today, for the majority of media organizations, it is almost taken for granted that at least some of the infrastructure and services will be cloud-based, and what or how much is on-premise depends largely on legacy storage or systems that need to remain operational. Cloud-based platforms are already helping to reduce those start-up costs and diminish risk. With minimal initial investment, systems can be spun up and, especially in integrating other systems, sandbox and staging systems can be managed cost-effectively,” Schragmann explained.
“The cloud is bringing large economies of scale and this is transforming content supply chains everywhere. I expect that five years from now cloud-native supply chains will be the norm. With this evolution, we’re going to see business globalization and democratization, with the opportunity for companies of any size to operate at a global scale. I’d also like to think that our industry can move closer towards industrialized workflows, with increased standardization of operational practices working to pre-set templates versus laborsome customized workflows. That would make incredible business sense,” said McMullan.
“With the expansion of cloud storage technology and continuously increasing storage tiers, there is a growing need for an abstraction layer, to manage the tiers automatically and provide media access in the most cost-efficient manner,” Mueller said.
“Serverless cloud is a new concept that is fast gaining traction with services like Amazon AWS Lambda and Microsoft Azure Functions, which free cloud users from fixed compute, storage, and bandwidth environments. Instead, serverless cloud environments invisibly scale on-demand as applications require. In truth, the servers are still there, with the technology acting as a layer of abstraction between the user and the platform orchestrating the compute and storages environments based on application demands,” Admans noted.
“A cloud-native supply chain model is certainly the overall trajectory, but the speed of adoption will vary according to region, local factors and appetite for change. This is why a flexible deployment architecture and well defined migration paths which take into account for example disaster recovery scenarios and cloud cost control are key to helping our customers on that journey,” said Powell.
“The trend in the industry is headed in multiple directions depending on the nature of the business requirements of the customer. We see customers that depend heavily on local talent accessing on-premise infrastructure, while others require geo-dispersed and WFH employees accessing systems quickly and easily. As more cloud-native services such as compute and high-performance storage along with point-to-point connectivity increases, we see an ecosystem where the customer has a choice of performing most if not all the on-premises workflows in the cloud based on the business goals they focus on,” Dabul said.
How will this further impact broadcast workflows?
“Increasingly, broadcast supply chains will migrate to the cloud, from ingest through delivery. For many, the goal is to push content directly into the cloud from the point of ingest or acquisition, and never have it come back to the ground until it reaches the viewer’s device. This kind of efficiency and agility will allow broadcasters to scale the number of channels and services they can offer dramatically,” Stedman said.
“It would allow content creators and distributors to focus more on creative tasks: plan, produce, manage, package and distribute their content quicker, helping them stay competitive in a content-hungry world,” said Zarouk.
“Just as new storage options are introduced often, changes to storage technology will also become more common. It’s reasonable to expect to have multiple Cloud vendors at single installations. There is no cloud vendor that provides a storage solution outside of their own technology. AWS will not manage a Google cloud. To maintain a seamless operator experience, no matter the shuffle of media in the background, it is necessary to have a content manager which can completely abstract the storage,” Mueller told us.
“Cloud-native supply chains continue to have a positive impact in the overall broadcast world. Workflows are no longer tied to specific facilities both in terms of hardware/software accessibility and access to the right creative talent. The flexibility that comes with hosted solutions allows for the creativity required to reach niche audiences on specific devices and platforms. As a result, content creation volume is at an all-time high,” Young said.
“The commitment to sustainability has a massive impact on downstream energy as the media and entertainment industry transitions to the cloud. Many broadcasters require suppliers to commit to sustainability with specific key performance indicators (KPIs) as part of their vendors’ contractual obligations. The most significant improvements come from suppliers who have transitioned to the cloud,” said Admans.
“Tech-savvy broadcasters who are looking to seriously take advantage of what the cloud and cloud-based workflows have to offer are waking up to the difference between solutions which have been lifted and shifted to the cloud versus products which have been designed from the ground up with cloud in mind. This applies to both the commercial opportunities for cost saving as well as the actual operational workflows which can be achieved,” Powell pointed out.
“As cloud-native services become more cost effective we see a future where broadcast ecosystems can be completely hosted in the cloud and purchased as flexible business models to ensure effective ROI,” said Dabul.
Geoff Stedman, SDVI
Russell Vijayan, Digital Nirvana
Karsten Schragmann, Vidispine
Mathieu Zarouk, Dalet
Savva Mueller, Telestream
Rick Young, LTN Global
Bill Admans, Ateliere
Camilla Powell, Grass Valley
Alan Dabul, Ross Video
Eric Newbauer, Studio Network Solutions