Microsoft to bring machine learning for video to Azure Media Services

Back in September, Microsoft pursued its efforts in media cloud streaming by expanding Azure Media Services with new services, certifications, and partnerships. But according to a new report from Techcrunch, Microsoft will soon improve its cloud services for premium video workflows by leveraging on its machine learning technologies.

Microsoft detailed all the new features at the annual NAB show in Las Vegas today. First, Azure Media services will get an automatic video summarization feature through analysis of the most interesting parts from a video. Next, Media Services will also benefit from a new speech-to-text indexer that can understand eight languages and an optical character recognition feature that can analyze text displayed in videos. More, the company is also leveraging on Microsoft Cognitive Services (the same team that worked on CaptionBot, an AI online tool with picture-recognizing abilities) to bring face recognition features to detect people’s emotions.

Cognitive-Face-01

Azure facial recognition.

Finally, Microsoft has also built a new Hyperlapse technology right into Azure Media Services for stabilizing and time-lapsing videos. According to Techcrunch, this feature was already available as a public preview but Microsoft just removed a previous technical restriction that limited users to videos with up to 10,000 frames.

All these new features could be very useful for media companies currently adopting cloud-based strategies. We’ll stay on top of this to see if Microsoft continues to gain momentum as a media solutions provider in the following months.

Share This
Further reading: , , , , ,