More organizations and enterprises are embracing video as the preferred medium to train their employees, engage their customers and document business functions. 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. Azure Media Analytics services are built using the core Azure Media Services platform components and hence are ready to handle media processing at scale on day one.
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.
Veritone, which provides cognitive media solutions aimed toward developing intelligent cloud solutions, has built and deployed near-real time intelligence solutions for broadcasters, advertisers, media firms and other enterprises via its Cognitive Media Platform. As an early adopter of best in breed “cognitive engines”, Veritone has already seen tangible business benefit from incorporating Microsoft's Azure Media Indexer service onto its platform. Now part of Azure Media Analytics,Azure Media Indexer had helped Veritone to quickly analyze and transcribe audio and video content in multiple languages for its customers securely and efficiently at scale.
Veritone has already begun implementing additional Azure Media Analytics services in private preview as it expands its cognitive engine portfolio. Ryan Steelberg, President of Veritone Media and Co-Founder of Veritone, Inc., said, “Organizations and enterprises today are producing video content at an incredible rate. As a company dedicated to building intelligent cloud solutions across industries, we’re excited to incorporate Microsoft Azure Media Analytics’ advanced machine learning technology in speech and vision onto our platform. This integration advances our cognitive capabilities, to further enable our customers to uncover actionable insights, currently deployed across the broadcast, advertising, law enforcement, and legal industries.”
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.
As part of this announcement, Azure Media Indexer (generally available since September 2014) and Microsoft Hyperlapse for Azure Media Services (which is available as a limited preview media processor) will join the collection of services under Azure Media Analytics. All services in Azure Media Analytics with the exception of Azure Media Indexer are available at no cost for a limited time. You can check out the demo page for Azure Media Analytics to look at the possibilities of these technologies first hand.
In Detail: Azure Media Analytics
Starting today, customers who use Azure Media Analytics will have access to the following services:
-
Indexer – With this announcement, we are introducing support for six new languages (eight total) for speech-to-text functionality via a new media processor called Azure Media Indexer 2 Preview. The full list of supported languages includes: English, Spanish, French, German, Italian, Chinese, Portuguese and Arabic. Support for additional languages will be available in the coming months.
-
Hyperlapse – Microsoft Hyperlapse is a result of over 20 years of computer vision research at MSR, combining video stabilization and time lapsing to create quick, consumable, beautiful videos from your long form content. Since the free public preview release of Hyperlapse, we have removed the limit of 10,000 frames. Besides creating time lapses, you can also use Hyperlapse to create stable videos from shaky videos captured via cell phones and camcorders.
-
Motion detection – As the name indicates, you can use this service to detect motion in a video with stationery backgrounds. This is ideal for customers who want to check for false positives on motion events detected by surveillance cameras on the surveillance video feeds.
-
Face detection and Face emotions – Using this service, you can detect people’s faces and their emotions, including happiness, sadness, surprise, anger, contempt, fear, disgust and indifference/neutral. This has several useful industry applications, described below, including aggregating and analyzing reactions of people attending an event.
-
Video summarization – Video summarization can help you create summaries of long videos by automatically selecting interesting snippets from the source video. This is useful when you want to provide a quick overview of what to expect in a long video.
-
*Video Optical Character Recognition – As the name indicates, you can use this services to extract text that is displayed in the videos. This can be very useful in building a smart search engine for videos which has a lot of text. The search engine can index not just the title and description but also the text extracted from the videos.
-
*Content moderation – This service is designed to cut down the manual effort involved in reviewing content for offensive, illicit and inappropriate material and is ideal for you, if you are dealing with a lot of user generated content submitted from anonymous sources.
*Video Optical Character Recognition (OCR) and Content Moderation are available to our customers in private preview. In order to gain access to Video OCR and content moderation, email: amsanalytics@microsoft.com.
Scenarios
Below are a few example of the scenarios where Azure Media Analytics can help organizations and enterprises across industries glean new insights from video to create more personalized audience and employee engagements, as well as more effectively manage the sheer volume of video content
-
Call centers – Even with the advent of social media, customer call centers still facilitate a large percentage of customer service transactions. Encoded in this audio data is a wealth of information about customers that can be analyzed to improve product roadmaps and also train call center employees to achieve higher customer satisfaction. By using Azure Media Indexer, customers are able to extract text and build a search index and dashboards to extract intelligence around most common complains, source of complains and other such relevant data.
-
Education – – In the educational sector specifically, we often see PowerPoint slides displayed in videos. By using the video optical character recognition service in conjunction with the Indexer, customers can extract both spoken works and displayed sentences and build a much richer search index which can help students find relevant videos much quicker.
-
User generated content moderation – From news media outlets to police departments, many organizations have public facing portals where they accept UGC media, such as videos and images. The volume of content can spike due to unexpected events. In these scenarios, it is near impossible to conduct an effective manual review of the content for appropriateness. Customers can rely on the content moderation service to focus on the content that is appropriate.
-
Surveillance – With the growth of IP cameras, there is an explosion of surveillance videos. Manually reviewing surveillance video is time intensive and prone to human error. Azure Media Analytics provides several components such as motion detection, face detection, video OCR and Hyperlapse to make the process of reviewing, managing and creating derivatives easier.
We’re truly excited to enhance our intelligent cloud with the addition of Azure Media Analytics. If you’re attending the NAB Show this week, we’d love for you to stop by our booth at SL6810 to learn more. Also check out the demo page for Azure Media Analytics to look at the possibilities of these technologies first hand