Artificial Intelligence: Changing the Game of Media Asset Management (MAM)
PFT Blog Team | 17 Nov 2017

Artificial Intelligence: Changing the Game of Media Asset Management (MAM)
Artificial Intelligence: Changing the Game of Media Asset Management (MAM) Click To Tweet

By Animesh Swain
Product Manager, Product Line Management

Artificial Intelligence (AI) is today driving business transformation across literally every industry vertical. Self-driving cars, virtual personal assistants and warehouses run by AI-powered robots have become popular realities. The Media & Entertainment (M&E) industry has been buzzing about AI for years, and as the business benefits become more transparent, harnessing AI innovations is fast becoming top priority for broadcasters, studios and digital organizations alike. Currently, several tech giants like Microsoft, Google, IBM and Amazon deliver several broad spectrum AI innovations for M&E players, while others like Limecraft and GrayMeta deliver AI technology for specific content operations.

What makes AI a golden ticket for media companies is not just its ability to automate tasks like translation, but also its potential to generate automated metadata for enhancing searchability and discoverability of content. This in turn plays a major role in increasing monetization of assets and achieving lowest Total Cost of Operations (TCOP).

Typically, AI capabilities in the M&E sector include:

Audio Analysis:

  • Speech-to-text: Converting spoken video recordings into readable text
  • Language Translation: Translating text from one language to another
  • Sentiment Analysis of speech: Determining the attitudes, opinions and emotions expressed

While capabilities in the first two areas exist today with workable accuracy, achieving high levels of accuracy for Sentiment Analysis continues to remain a challenge.

Visual Analysis:

  • Object recognition: Identifying multiple objects within a video (including humans, cars, animals etc.)
  • Action recognition: Identifying how these objects interact with one another (singing, driving, running etc.)
  • Emotion recognition: Identifying emotions expressed while performing these actions
  • Face recognition: Identifying presence of individuals (including celebrities) in a video based on a library of known faces
  • Logo recognition: Identifying specific companies based on logo/brand imagery in a video

The classification above has been done to identify specific and relevant AI information elements, which are discoverable today using off-the-shelf AI technologies. These specific elements form key information building blocks that help drive efficiencies in day-to-day media workflows. For instance, a Promo Production team tasked with creating movie promos for a network will want to search for specific, high intensity scenes that depict the protagonist performing certain actions (car chase, shooting, fighting etc.). An AI-enabled MAM system can help retrieve this information at lightning speed. Moreover, the MAM’s Clip Edit tools can help refine these clips before sending EDLs to the edit table.


Visual Analysis: Elements and Automated Discovery using AI

This is just one example of the tremendous value addition that an AI-enabled MAM offers. Given below are a few more specific use cases where this technology can be leveraged to reduce expenses and tap new revenue opportunities:

Subtitling/Closed Captioning

Audio analysis of an asset using a speech-to-text engine can help generate a transcript with reasonable levels of accuracy. While it is a known fact that speech-to-text technology works best only for certain types of content, such technological interventions in the Localization process have the potential to bring down the overall manual effort to just QC. Moreover, when auto-generating subtitles from transcripts, AI driven spotting tools can identify precise time-codes using standard rules like reading speed, line breaks etc. To ensure high quality output, tools that highlight low confidence areas in the auto-generated script can be used, as this helps professionals decide whether to use or discard the AI generated output. Also, for the translation of scripts into different languages, sophisticated tools are now available to perform automated language adaptation and translation.

Creating Compelling Content from MAM/Archive

A MAM enriched with AI metadata makes it easy to search through a vast content repository for topics and archival footage related to a particular theme in order to create a riveting story. Such metadata driven research is particularly useful when creating news stories or documentaries from archives that generally have limited levels of tagging. Here, speech-to-text conversion can be undertaken to tag all archival content with AI generated metadata, thus making the repository easily searchable. Additionally, content can be auto tagged using Object Recognition as well as Face/Location Recognition technology to further enrich metadata and build highly relevant stories around specific celebrities/public figures/themes etc.

Identifying Brand Imprints

A brand owner sponsoring a particular TV show/sports series can now seek the precise number of brand imprints in a given match or series. An AI-enabled MAM helps automate such tedious tasks, thereby saving time and reducing costs. Today, advanced analytics capabilities are also available to identify high value brand Imprints (for instance pin-pointing brands displayed in the clip when Lionel Messi scores a goal) by performing searches across AI generated and manually logged metadata.

Monetization of Stock Footage

With Object Recognition, AI makes it easy to discover elements like landmarks, locations etc. within stock footage, thereby boosting monetization.

Mastering Operations

An AI-enabled MAM can search for specific clips in a video that often need to be deleted for Compliance purposes. Typical examples include nudity, drug abuse, smoking, violence and abusive text on screen. Edit Decision Lists (EDLs) generated from such clip lists can aid editors in quickly zooming in on the specific areas of interest while performing Compliance edits on high resolution assets. Such AI-led EDL generation helps increase operational efficiencies and achieve faster time-to-market. However, more is still desired from an accuracy and spectrum perspective in this arena, as identification of such ‘Not Safe for Work’ (NSFW) segments is often not enough. In another use case, AI-led Object Discovery can help find elements like ‘color bars’ and ‘blacks’ within a video, and an AI-enabled MAM can automatically remove these, thereby streamlining re-mastering operations.

The Roadmap Ahead

Most businesses today have started experiencing the impact of vast improvements being made in the domain of AI. The race to embrace AI is on, and early adopters have already begun reaping a host of business benefits which include enhanced efficiencies and increased monetization. For M&E enterprises struggling to cut through the clutter of video explosion, leveraging AI to ensure easy discoverability of assets has become a critical need. In order to win, companies must adopt an AI-enabled MAM, driven by a powerful search engine, and augmented with diverse AI capabilities. This forms the key foundation for driving automation-led efficiencies across the content supply chain, slashing costs and improving the bottom line.

Click here and we'll get you a meeting with our Subject Matter Expert


Subscribe to get emails delivered to your inbox

Email *

Thank you!
Our team will get in touch with you shortly.