We’ve all had the dreadful prospect of watching an hour-long video of a presentation, lecture, or corporate “fireplace chat.” This trend has only been exacerbated by the COVID-19 lockdown, which has turned many of us into blurry homeworkers and learners.
Companies and colleges produce hundreds of millions of hours of elongated video every day, but because the costs of editing and producing it can be exorbitant, most of us end up getting a lot less value than we could otherwise.
Created to address this paradox, Educational Vision Technologies (EVT) has developed a number of services that provide university students and knowledge workers alike with more digestible versions of elongated video content, using machine learning algorithms to break down long videos into short videos chapters of only a few minutes each, accompanied by full transcripts and other notes.
Originally developed with disabled students in mind, the service also benefits disabled students who are either unable to attend class or who find the note-taking assistance useful, as well as knowledge workers and other remote workers. Monal Parmar, founder and CEO of EVT, notes:
Studies have shown that it takes more cognitive effort to take notes while listening to a lecture than to play chess. It makes no sense for students to use their cognitive bandwidth too much to write everything from the whiteboard or blackboard. Providing notes gives students the flexibility to take as few or as many notes as works best for them.
Research from Kansas State University confirms the idea that students who “use externally provided notes… generally achieve more on exams than students who review their own notes.”
At the students’ own pace
The EVT service is used by some departments of the University of California San Diego (UCSD), as well as a handful of other universities and a professional training organization.
The technology allows students with disabilities to “participate in an academic lecture multiple times and at the student’s individual pace,” Joanna Boval, director of the agency for students with disabilities at UCSD, said in a letter.
EVT’s Parmar says the company has raised $700,000 and currently has four full-time employees and three contractors.
The company, which has built its service on Oracle Cloud Infrastructure (OCI), is also participating in the Oracle for Startups program, which he believes has led to a number of sales opportunities. He explains:
Many cloud accelerator programs are simply referred to as accelerators in their name, but do little more than provide cloud credits and some technical guidance. Oracle does much more than that.
Powered by machine learning
EVT offers two variants of its services. EVT Bloom allows customers to upload their own videos for management, while EVT Learning Systems uses an on-premises device.
EVT Bloom uses machine learning to split recordings into short video chapters and automatically creates an interactive table of contents, searchable speech transcript, speaker summaries, and quiz questions. The titles of the short videos are set as headlines on EVT’s web platform so that they can be spoken by a screen reader to help people with visual impairments better navigate through the video content.
Parmar says the company had to overcome some machine learning challenges and figure out how to scale the service. It also needed to troubleshoot hardware errors on the on-premises device used by EVT Learning Systems, such as overheating, as well as unreliable power supplies, network issues, and supply chain disruption. In contrast, the software offerings were more streamlined to develop. “We simply moved many of our core algorithms to the cloud and turned them into microservices,” he says.
EVT Bloom plans range in cost from $33 per hour to $60 per hour of video billed on an annual basis. According to Parmar, AI curated content of one-hour lectures is available within 30 minutes, and fully corrected versions within 24 hours.
The COVID-19 pandemic accelerated adoption of the technology at UCSD, but the university had already used it to advance its mission to provide equitable and inclusive education for a diverse student population. In a statement, his chancellor Pradeep Khosla noted:
They have not only helped us improve accessibility, but also improve student satisfaction for distance learning.
Oracle comes through
Processing for EVT Bloom is on OCI, with the content encrypted and stored in OCI Object Storage. Parmar explains:
We developed our machine learning microservices on OCI. During many meetings, Oracle engineers guided us through the prototype and development of our machine learning microservice infrastructure in the Oracle Cloud. For any technical challenge, Oracle engineers were on hand to support us and regularly check how things are going.
Parmar notes that cloud costs may have been a factor for his business, as EVT has thousands of how-to videos streaming from its website. But it was able to achieve “significant cost savings” on video streaming because the Oracle service doesn’t charge the first 10 TB per month for outgoing traffic.
Parmar received a bachelor of science degree in electrical and computer engineering from UCSD in 2016 and was working toward a master’s degree with a focus on machine learning while working on his startup when the pandemic hit. “Something had to be given,” he says.
He plans to return to graduate work in 2023 or 2024 – and could take advantage of his own brainchild to complete his studies.