Microsoft Project Oxford Natural-Language Machine-Learning Beta Debuts

By Pedro Hernandez  |  Posted 2015-10-27 Print this article Print
Microsoft Project Oxford

More developers can now take the machine-learning technology behind Microsoft's viral hit,, for a spin.

Developers looking to build intelligent apps that understand natural language no longer have to wait for an invitation to participate in the Project Oxford private beta.

Project Oxford is Microsoft's collection of machine-learning application programming interfaces (API) and services. During this year's Build developer conference, the technology helped the company score a rare viral hit with its age-guessing site, Within hours of being launched, the site garnered hundreds of thousands of image submissions as users tested the site's attempts to figure out the ages and genders of photo subjects, sometimes to comedic effect.

"Starting today, we've further expanded the reach of Project Oxford's Language Understanding Intelligent Service (LUIS) with public beta availability, Chinese language support and more pre-built models," Ryan Galgon, senior program manager, Microsoft Technology and Research, said in an Oct. 26 statement. "In addition, beta versions of the Project Oxford face, computer vision and speech SDKs are now available on GitHub."

Formerly available on an invitation-only basis, the public beta of LUIS enables developers to build apps that accept natural language inputs, similar to the way Cortana, the digital assistant found in Windows 10, can be used to look up sports scores or set reminders by uttering commands in plain English. Now those capabilities are being opened up to developers targeting the Chinese market, revealed Galgon.

"Chinese support has been one of our most commonly requested features, and now you can create LUIS applications in English or Chinese. In addition, LUIS correctly processes Chinese utterances that include fragments of English," he stated.

LUIS also fits better into certain application development workflows.

"Now, all of the data you've entered into a LUIS application can be downloaded to a JSON [JavaScript Object Notation] object and new applications can be created by importing that JSON object," Galgon said. "This allows developers to copy applications, share applications with others and check their applications into source control—for example, to be versioned alongside the code for the client app that calls LUIS."

Microsoft also has published more pre-built LUIS models, fashioned after the code the company uses for its own products. Galgon reported that his group has also "more than tripled the number of intents and entities," from 56 to nearly 200.

In a further expansion of Project Oxford's reach in the developer community, Microsoft published its face, computer vision and speech software development kits (SDKs) to the popular GitHub code repository.

"This will enable us to better work with the community and build a more inclusive, robust platform," Galgon asserted. "The repo contains Windows and Android support for face, computer vision and speech APIs, and we are working to on-board the remaining packages as well as increase our SDK offerings in iOS and Node.js, both top requests from developers." Alternatively, interested developers can also access Project Oxford components for Windows and Android on NuGet and Maven, respectively, he added.


Submit a Comment

Loading Comments...
Manage your Newsletters: Login   Register My Newsletters

Rocket Fuel