Microsoft gives developers access to its image, speech recognition APIs

Microsoft's offers developers access to facial, image and speech recognition API's.

By Silky Malhotra Published Date
06 - May - 2015
| Last Updated
06 - May - 2015
Microsoft gives developers access to its image, speech recognitio...

Microsoft has released new machine-learning APIs in beta, which can calculate a person's age based on their photograph. Microsoft demo under its Project Oxford program went viral a day after it was released. The tech demonstrates how developers can upload photos of faces and the system automatically deciphers the age of the person in the photo.

The new APIs allow developers to add face detection and recognition features to their apps and will find out what the user’s age is automatically. According to reports, uses some of the new developer services that are a part of Project Oxford. The service is based on modern deep learning techniques that Microsoft has worked on over the last few years. The API supports face detection in images, the ability to find similar-looking faces as well as face verification to check whether two faces belong to the same person. It has a vision API with optical character recognition capabilities for creating smart image crops that put the subject into the center of the cropped images.

It also has a speech recognition API that helps developers understand their user’s intent over time. The API offers speech-recognition services for speech-to-text conversion as well as a text-to-speech service that turns written text into audio. All the three services are now available as a public beta. Apart from this it also has a fourth API that lets developers build custom language understanding into their applications. Microsoft has made the services available for free.

Ryan Galgon, a senior program manager on the Oxford project at Microsoft Technology and Research, stated at the Microsoft’s Build developer conference that the Project Oxford and the age-detection project was the result of a collaboration between different groups inside of Microsoft. Rival Google have also unveiled an artificial intelligence system called FaceNet that according to company claims is accurate 99.6% of the time. The technology will help in searching people as well as automatically tagging photos. The feature will also make it easier for web companies to analyze global trends and their users' social networks. Apart from Microsoft and Google, social networking giant Facebook as well as Yahoo have been investing in developing deep learning techniques.

Source: Project Oxford