Do you know what was the rant, a couple of months ago, about this picture?

Computer Vision example image

Or do you wonder how How-Old.net that calculates (correctly or not) your age works? If you are interested in – the read on.

The issue with the picture was not in the picture itself but with the alternative description that could be seen in HTML was on Facebook. Yes, that’s right the picture was originally posted on Facebook. What was in the description?

Image may contain: 26 people, people smiling, people sitting, people standing and indoor

That’s pretty accurate accurate description if you ask me. And of course it was not entered by hand. It was done by the machine.

Microsoft Cognitive Services

Here’s where aforementioned services comes into play. Those services are set of APIs that can be used to automate such tasks as image analysis. There are 5 main categories that we can take advantage of the add a bit of intelligence to our applications.

Vision

Focuses on image & video analysis. We not only get information what is in the picture but we can also ask about face detection or even to get information about people’s emotions. If you are worried that you might open image NSFW during work, run it by the service and check racy & adult indicator. If they are above certain threshold it might be safer to wait with opening that link till after hours.

Speech

Here’s the standard text to speech functionality but also speaker selection (detecting who’s speaking from the group of known ones) and speaker validation – checking if this is the voice of the person that claims to be.

Language

LUIS that stands from Language Understanding Intelligent System is here the more interesting  choice for exploration. Services that allows you to understand text and act accordingly to what is your user/customer/etc. writing.

Knowledge

Mostly academic paper exploration, Q&A maker and recommendations.

Search

Standard set of API to allow searching the net for text, images, news and videos. Nothing fancy here with a little bit of auto suggestion functionality.

 

Looks like we have a nice set of APIs that might enhance applications with a little bit of “human” factor. I would definitely want to use them a bit more.

Have you heard about Microsoft Cognitive Services before? Or maybe you’ve already using them? I would love to hear your story.

Founder of Octal Solutions a .NET software house.
Passionate dev, blogger, occasionally speaker, one of the leaders of Wroc.NET user group. Microsoft MVP