Microsoft is taking autocorrect to the next level

Microsoft is using artificial intelligence and Windows Machine Learning (ML) to improve its products, including Office 365. During the third Build keynote, corporate vice president of the Windows Developer Platform Kevin Gallo used Microsoft Word as an example, stating that the company’s goal is to make everyone a better writer. How? Through grammar checking powered by Windows ML and artificial intelligence. 

“Some areas are very, very hard to detect with traditional algorithms,” he said. “For example, you get into a car, but onto a train. There is a shadow on the road versus there is fog on the road.” 

He said this problem is personal due to his daughter, Anna. He said she’s struggled with grammar her whole life and felt that she would never be a good writer. Her woes are understandable: The English language is complex, with words having different meanings depending on the subject. His example of a car versus a train is another good point. 

But by incorporating machine learning, Microsoft can help pinpoint these small errors. In a demonstration, he produced two identical documents: One edited using the current version of Word, and another edited using Word powered by machine learning. The standard version didn’t pick up on three missing determiners while the prototype Windows ML-powered version highlighted the three nouns that were missing their determiners. 

“We’ve trained the grammar checker and it now can suggest corrections that I can take action on and fix,” he said. “We’re running this on Windows ML, which enables Word to build an experience that is low-latency, has high scalability because there are a lot of Word users out there, and it can work offline.” 

The big news here is that Microsoft’s products, such as Word, are now relying on machine learning algorithms running locally on a Windows 10 device, and not in the cloud. On devices that support DirectX 12, Windows ML can utilize the PC’s graphics chip while using the processor as a secondary resource. And because these algorithms are running locally within apps installed on a device, the results are extremely quick. 

Group program manager Kam VedBrat introduced the new Windows ML application programming interface (API) in March, a platform that enables developers to implement pre-trained machine learning models in their apps and experiences. On-device A.I. is necessary when there is poor or no connection to the cloud, if the data is just too large, or if customers simply don’t want their data uploaded to the Azure cloud. 

VedBrat’s demonstration included an app that can scan circuit boards for problems. The machine learning model was developed in the cloud (Microsoft Azure) using labeled images of boards known to be working or defective. Thus, the app can either pass or fail a circuit board after scanning for the proper components. The driving point is that the evaluating process can be done locally on the Windows 10 device, and not in the Azure cloud. Even more, the PC relied on its graphics chip and DirectX 12 instead of the processor. 

Editors' Recommendations