Not long after Google rolled out an early version of its Cloud Vision API tool for image analysis, Microsoft has introduced a preview version of Azure Media Analytics. As with Cloud Vision, the new service should have significant implications for enterprise applications as companies make broader and better use of images and video to help train employees, collaborate and connect with customers. Here are some of the key details of Azure Media Analytics from an official Microsoft blog post:
Cloud computing no doubt makes it effective to store, stream and access these large media files, but as companies grow their video content library, they must have an equally effective means for extracting new insights from video in order to create more meaningful, personalized interactions with their audiences and take their business to the next level.
To address this growing need in the marketplace, the Azure Media Services team is proud to announce at NAB Show 2016 the public preview of Azure Media Analytics, a collection of speech and computer vision services at enterprise scale, compliance, security and global reach.
We are introducing these services based on feedback and requirements from customers across industries, from public safety, government and surveillance to retail, education, automotive and more. With Azure Media Analytics, we aim to improve productivity by cutting down the manual effort required to manage, process and review high volumes of video.
Just two weeks ago at Microsoft’s Build Developer Conference, we showcased some of these capabilities as part of Cognitive Services, enabling developers to quickly get started with vision capabilities for video at limited scale and bring this advanced functionality into Bots and applications. Azure Media Analytics brings these video APIs to enterprise environments with the full scale, compliance, security and global reach required by large organizations.
Microsoft is initially supporting eight languages in the service, with more to come.
Azure Media Analytics tackles video analysis from a number of directions, such as with Hyperlapse, which provides video stabilization and time lapsing for creating shorter, more effective videos from large amounts of footage; motion detection, which could be used by "customers who want to check for false positives on motion events detected by surveillance cameras on the surveillance video feeds"; optical character recognition for extracting textual data from videos.
Perhaps most interestingly, facial and emotion detection capabilities for determining the emotions people are feeling in videos. It's easy to see how this capability could be applied to a variety of enterprise applications, from videoconferencing software to productivity tools such as email—if an application can tell whether a user is overwhelmed, it could conceivably reorder their inbox or hide unnecessary messages, as Constellation Research VP and principal analyst Alan Lepofksy notes.
For its part, Microsoft cites a number of potential use cases for Azure Media Analytics, including call centers, education, content moderation and surveillance.
How effective the service proves to be remains to be seen. But the general idea of it is welcome, Lepofsky says.
"Video has gained popularity as a key method of sharing information and engaging with colleagues or customers," he says. "It can be used for product demonstrations, training, review of products or events or just as way of sharing thoughts and asking questions in a more visual manner than just text provides. However, one of the drawbacks to video has always been that it’s hard to search, index and summarize."
"With a blog post you can quickly skim the article, but it’s hard to do that with a five, 10 or 20 minute video," he adds. "By applying AI services such as speech-to-text, automatic indexing and facial recognition, Microsoft is helping improve the way people will be able to interact with video content.”
Reprints
Reprints can be purchased through Constellation Research, Inc. To request official reprints in PDF format, please contact Sales.
