Monday, 24 August 2009

Websites that use the spoken word will empower the illiterate

THE internet, wonderful though it is, reinforces one of life’s fundamental divisions: that between the literate and the illiterate. Most websites, even those heavy with video content, rely on their users being able to read and—if interactive—write. Building your own site certainly does.

Guruduth Banavar, the director of IBM’s India Research Laboratory, wanted to allow people who struggle with literacy to create websites. So he and his colleagues have devised a system based on what is known as “voice extensible markup language”, a cousin of the hypertext markup language used on conventional websites, that allows a website to be built and operated more or less by voice alone.

Dr Banavar thinks mobiles could be made to work much harder. His voice sites are hosted on standard computer servers and behave much like conventional websites. At their most basic they are designed for local use, acting as portals through which people can find out such things as when the mobile hospital will next visit their village, the price of rice in the local market and which wells they should use for irrigation. Instead of typing in a web address, the user rings the website up. Then, with a combination of voice commands and key presses, he navigates through a spoken list of topics and listens to subjects of interest.
Voice Browsing: How Two Great Ideas Go Great Together
Voice browsing technology is a rapidly-growing field. Whether or not it proves to be the next internet, it deserves a careful examination in its present form. The programming language responsible for connecting voicemail, live agents, and speech enabled sites is called VoiceXML, which was devised by Lucent, IBM, Motorola, and AT&T. IBM has several patents on its natural language understanding engine which uses probability to guess what people mean if the words are unclear.