VizWiz phone app provides virtual vision to the blind

There are so many things that a normal person would take for granted, for instance wearing coordinated clothes or picking a favorite brand of a bag of chips, however these everyday tasks become demanding for someone who is visually disabled, often leading them to rely on someone restricting their independence. To solve this, a team of researchers from various universities including the University of Rochester have designed a new app available on the iPhone, called VizWiz. It's an iPhone application designed for use with the VoiceOver screen reader included on the iPhone 3GS.

vizwiz phone app for blind

When users start VizWiz, it starts to ping its remote server to indicate that a question may be asked soon, so that the server can start recruiting workers if necessary. Once the picture is taken and the question is asked, VizWiz sends these to the remote server. The picture is compressed on the phone through a background process, while the sound is recorded to reduce the latency required to send the image. Sound is recorded in MP4 format using hardware compression on the phone. When the server receives the question and image, it calls a speech recognition service (currently the Windows Speech Recognition Engine) to convert the question to text, and adds the image and question to its database. Importantly, speech recognition does not need to work because workers can also listen to the original question in an included Flash player.

With this web based program providing them virtual vision, the blind will now be able to read menus, identify canned foods, locate things in their homes, coordinate clothes comfortably and even frisk out for an empty bench in a park without having to walkover.

Google’s Android platform and the Apple iPhone 3GS now include free screen readers. The iPhone has been particularly popular among blind users, which motivated the developers to concentrate on it for VizWiz. Apple’s stringent controls on the applications available on its online store and tighter integration of its screen reader (VoiceOver) with the operating system has resulted in a large number of accessible applications.

Via: New Scientist