There are hundreds of millions of people around the world who have physical disabilities that prevent them from using a mouse or keyboard. In 2011, the UN declared internet access as a basic human right and the World Wide Web Consortium has made considerable efforts in improving accessibility to the internet—all without much progress.

We wanted to build a solution that could use existing technologies and fit seamlessly into the existing structure of the World Wide Web. In the past, speech recognition has been used for transcribing text and basic operating system commands, but there hasn't been any significant progress into developing a true solution that integrates speech with web navigation. We solved this problem with Lucy.

Lucy enables you to navigate the web hands-free with natural language. Ask Lucy to "go to CNN", "click on the article about Tesla's new car," or even "search for HackPrinceton." Open and close tabs and scroll up and down web pages.

We solved several technically-challenging problems with Lucy. One of Lucy's best features is her ability to discern specific parameters from any phrase and to identify a user's intent even from the most basic request. For example, Lucy is able to find the relevant news article on the New York Times website from "show me the article about Trump's debate last night." Using machine learning, we built Lucy to accommodate the endless variety of requests that a user can give her.

Lucy provides internet accessibility to the tens of millions of people around the world living with a physical disability. Our vision for Lucy is an internet browser that provides full functionality through speech recognition.

Share this project:
×

Updates