With a new development called Life Book by a group of students from Kasetsart University and Chulalongkorn University, people can now enjoy reading books or any text document more easily and with more fun.

The team, who are all senior computer-engineering students, has developed a program to reduce the illiteracy problem among Thai people. Especially for some people who can't yet read, they said the program would help them understand what the book or text is all about, just by listening.

Jatuporn Sukkasem, a senior student from Kasetsart University who is one of the development team, said the Life Book program was initially designed to support only English language. The team adopted optical character recognition (OCR) technology, which helps convert scanned documents into text, combined with text-to-speech technology to turn text from paper-based documents into sound.

The program recently won an award at Imagine Cup 2007, a software competition hosted by Microsoft Thailand.

"We hope the program will help those who cannot read English get more understanding, as apart from the program helping to convert text into sound, it will also show the meaning of each word as a picture. This will even help people who are not fluent in English to understand the vocabulary," he said.

To use the program, people have to stay connected to the Internet. A Web camera is also required to capture text users want to read. Once captured, the program will turn it into the form of a document image then a normal text file using OCR technology before sending it to be processed through text-to-speech into voice.

The team developed the OCR using an open-source base as well as the text-to-speech software, Jatuporn said.

"Then we added special features to turn the words shown in the text into pictures so users will more easily understand the meaning of each word through pictures.

For example, if the text shows a word like "apple" or "tree", the program will display pictures of apples or trees to give users the word's meaning.

Jatuporn said users would see a three-dimensional book page, specifically a paragraph, consisting of many pictures on top of many words, while their ears receive the synthesised voice to guide them on how to read this passage.

 "Users can see 3D graphics as the program was designed by Windows Presentation Foundation as a component of .Net 3.0. It helped us to develop the better graphics," said Jatuporn.

Apart from text captured through the Web camera, the program can also read human handwriting and then convert it to text files.

The team said they also developed the program to help users pronounce correctly.

For example, if there are three words such as "tree", "tea", and "three", the program will help users to know the different pronunciations and meanings by showing users the different pictures.

Jatuporn said the team also developed the program to support Chinese language and it planned to make it support Thai soon.

As the winner in Imagine Cup 2007, the team will get support from Software Park's Incubation Centre programme and receive training from Microsoft after graduation.

In August, they will represent Thailand at the World Imagine Cup 2007 in Korea.

Asina Pornwasin

The Nation