Bite-Size Video: How AI and Emerging Technologies are Transforming Video Production To Amplify Audience Engagement
As the consumption of video has increased over the last several years, broadcasters, sports teams, organizations have made concerted efforts to increase their content. And they’ve been largely successful--almost too successful, as many are figuratively drowning in video.
Now the question isn’t whether to video or not to video, but rather how to make the most of all that video--ranging from hundreds to thousands of hours in some cases--in an efficient manner. Shortform video has the ability to increase viewer engagement and drive higher monetization of content. But the challenge has been that the manual analysis of live streaming content to identify key highlights is both time-consuming and involves expensive manpower.
For a growing number of content developers, the answer is bite-sized videos created with the help of AI.
There’s a raft of benefits to bite-sized videos, primarily that they are easily consumed and shared, and that they are the preferred medium for many, including 25-34 year olds. Much of this is not new to content producers. What is new--and improving--is the video editing capabilities, driven by AI.
Smart Content Processing Using AI
Artificial intelligence (AI), machine learning (ML), deep learning, and natural-language processing (NLP) are enhancing video content production and streaming in major ways. These technologies will have an impact on streaming through all stages of production--from content creation, processing, and postproduction through consumption. AI is performing mundane, repetitive tasks in a fraction of the time and using minimum human resources than would otherwise be required.
Many of these improvements are the result of using AI and ML to perform metadata extraction and analytics, which was previously done manually. In one instance, which is typical, AI engine-based video content processing helped save 70% of manual editing time and achieved 80% cost-saving. It also improved efficiency by eliminating human errors.
Applications of AI in “Highlights” Streaming
“Highlights” are the fastest growing segment of video, whether in sports, movies, or television, with the business video industry estimated to reach nearly $20 billion by 2023.
AI can interpret streaming content and extract metadata by generating descriptive tags, categories, and summaries automatically. This helps in smarter analytics, content insights, and better content management. In sports, advanced solutions powered by AI and ML can identify specific game objects, constructs, players, events, and actions. This aids in near-real time content discovery and helps sports producers create highlight packages automatically even while a game is in progress.
One sports media production organization recently used AI to build interaction with its viewers on its app, providing relevant real-time content to them instantaneously during the broadcast of a major Grand Slam tennis event.
The AI-driven platform first helped identify the context of the content and map it across the broadcast. The AI identified and was able to tag nine key context data points. including aces, drop shots, double faults, emotional reactions. In all 490 hours of live broadcast content was analyzed, and 365 hours of key moments and highlights were identified and repurposed as smaller bite-sized content, saving 70% of manual editing time and achieving cost savings of 80%. The largest impact was the 120% increase in audience engagement.
It’s not just on a sports field. The power of AI in video content processing is being leveraged on the fashion runways as well. One of the world’s largest fashion video licensors, for example, was able to automate the process of generating timecoded metadata for thousands of hours of fashion show footage.
Through this effort, the firm enabled its licensees to easily search for individuals by name and metadata that previously was located across multiple systems. With the help of this new system, licensees are able to extract only the relevant clips directly from the video licensor’s web portal.
Obtaining Customer Insights and Business Intelligence
In the retail world, AI-powered retail analytics help enterprises obtain rich customer insights. Complex algorithms analyze and monitor large amounts of video and extract critical insights that help enterprises in business decision making.
For example, even the simple observations on how many customers enter a story and when they do so is a major analytic. It provides operational and branding insights, and reveals many other aspects of customer relationships. Ad analysis--video analytics--provides real-time and offline insights into the behaviors of customers interacting with ads and displays, helping increase sales and improve customer experience. These advanced video analytics allow stores to understand where and how long customers interact with products, displays, and the store space and helps stores in strategic placement of displays, checkout counters, and overall store layout.
The adoption of bite-size videos is expected to increase across various domains such as micro-learning, marketing, sports, fashion, and retail. Cutting-edge technologies such as AI, ML, NLP, and predictive analysis are amplifying audience engagement through revolutionary breakthroughs such as automated real-time creation of highlights and key moments packages, and advanced video analytics.
It will be a smart move for the companies that use video content to engage with their customers to include AI-powered solutions in their content strategy. This will help them achieve high audience engagement at a much lesser manpower cost, while significantly cutting down the production time, which directly translates as higher profits.
[Editor's note: This is a contributed article from Toch.ai. Streaming Media accepts vendor bylines based solely on their value to our readers.]