By: Rob Savette, VP Enterprise Market Development. Rob leads Ziotags enterprise sales and marketing initiative and brings 20 years of experience in managing and growing hardware, software, and services technology companies.
Going on a year since the pandemic took hold of our lives, there is no question that the work world has undergone a drastic transition. According to Upwork, 41.8% of the American workforce is currently remote, and an estimated 26.7% will still be working from home through 2021. Additionally, 80% of leaders surveyed by Gartner plan to allow employees to work remotely at least part of the time after the pandemic, and 47% will allow employees to work from home full-time.
On the whole, employees have become accustomed to remote working, and many of them actually prefer it to being in the office. According to a FlexJobs survey, 65% of employee respondents reported wanting to be full-time remote post-pandemic, and 31% want a hybrid remote work environment—that’s 96% who desire some form of remote work.
This new world world has manifested some new caveats however, one of which is the explosion of content creation on account of all work interactions being done digitally. Adapting the workforce to the requirements of automation, digitization, and other technologies has manifested a new set of challenges for businesses. One of the biggest challenges has been the escalation of video content. To date, many companies don’t have the capacity or technology to organize and streamline this content that has incidentally exploded, despite its miles and miles of potential.
What is Incidental Content Creation?
Incidental content is being created as a result of remote work. Many of our conversations are done online, and people are creating content via many different platforms without intention. Massive amounts of knowledge and data are being generated simply via workers’ day-to-day interactions. This endless data has the potential to be both useful to businesses and the employees that make them up.
Depending on how many meetings you have a day, whether it be via Zoom,Teams, or some other video platform – there’s a chance your company is recording it. Some companies store these recordings locally and others use the video platform to do so. For example, Zoom allows up to a gigabyte of storage on their cloud and then customers pay after that. The issue is, these self-made videos then simply sit in the cloud archive and collect dust with no future use in sight. This is a waste because those videos often have a massive amount of critical content, and could be used for multiple purposes within a company’s business needs.
The Challenge of Manual Video Tagging
So how do we make use of the videos, and then how can we share that useful information derived from them? Let’s look at two common ways to date in organizing video content – known as video tagging.
The first technique in tagging video media is by using editor tools, as seen in creative video editing applications. This follows a simplified process of crafting a title and then crafting tags that intuitively follow the content of the video. The second kind of video tagging is time-sliced video, which involves creating any number of minute intervals that can be tagged for focused context within the content of the video.
The number one issue with both of these methods is that many companies don’t have the staff, time, or technological capacity to apply these methods to their existing bank of video content. Additionally, people don’t want to go to ‘locations’ or ‘time slices’ – they want to find the specific content they need right then and there. Going to minute 15 is meaningless, but going to where an important topic was discussed that is needed in order to complete a task matters. Luckily, a new deviation of video tagging with this capability is emerging, and it is enhancing both the approachability and the scope of video organization and use.
Media Contextualization – Using AI to Change the Game
Through the use of Artificial Intelligence (AI) ,Optical Character Readers (OCR), and advanced transcription technology we are now able to achieve a high level of media contextualization.
OCR is the electronic or mechanical conversion of images of typed, handwritten, or printed text into machine-encoded text. OCR has been around for quite some time but they now have the ability to go even further and bring in the images of written context within a video and transcript it as well.
Media Contextualization simply means the ability to look inside the video and allow people to find and share, exactly what they are looking for. AI as applied to media contextualization delivers the ability to extract metadata that can be used to index, organize, and search your video content, as well as control and filter content for what’s most relevant.
Artificial Intelligence (AI) can take both video tagging and OCR further to create an all encompassing and comprehensive “media contextualization engine” through Machine Learning (ML) and Natural Language Processing (NLP) applications.
Machine Learning is the ability for AI to learn and adapt without following explicit instructions by using algorithms and statistical models to analyse and draw inferences from patterns in data. In the case of video, first the AI creates a transcript of the video’s audio content. Next, it captures the images in the video, and regurgitates and filters the visual context into the transcript via OCR. The AI’s capacity for ML allows it to draw conclusions from relational terms that help it to make a more clear and encompassing transcript.
Once the transcript has been created, Natural Language Processing comes in. With better understanding of the language being used by the participants in the video, the AI generates a natural language understanding report that consists of the category, concepts, emotions, entities, keywords, sentiments, top positive sentences, and word clouds. For example – the report might highlight specific uses of business jargon or metaphorical expressions.The more exposure the AI has, the more it’s understanding increases of the different vernaculars on how to communicate an idea. The more video you feed the AI, the more robust the transcript becomes. The more data and context the AI has, the more it learns. AI then becomes more comprehensive in it’s language understanding as it is given increased access to your company’s recorded meetings and conversations.
Using AI to create actionable business data
Many companies don’t understand exactly how AI in video applies to their particular operations. At the same time, they also have not been able to manage the massive amount of content they are creating, and how to share it across departments for absorption within their business’ infrastructure. Your company may need particular tags for a client’s product, or a need to distribute the contents of a meeting, or to share an interview of a potential candidate, or to streamline a video for all of the HR onboarding. The use scenario is huge, with multiple and diverse personas of need.
Microsoft CEO Satya Nadella noted in April 2020 that “we’ve seen two years’ worth of digital transformation in two months.” In order for the remote work world to continue by popular demand, companies must learn how to adapt the platforms they use most often so that they can continue to operate digitally. Video calling, recording, and content creation has grown exponentially (and sometimes incidentally) as a result of the remote work world. The good news is there is now a means to unlock the knowledge found within video content when AI is applied to the challenge.
Does your company have incidental video creators? Do you have a sitting bank of video content waiting to be unlocked? Respond with your feedback in the comments section below, and in our next blog post we will go into the specific ways in which AI can be applied to your business’ video content creation.