This is a research project aimed at developing mobile based dialog systems for information access in Urdu. The target user group is envisioned to be the semi and non-literate population of Pakistan, which faces the following multiple barriers including literacy (English literacy in particular) and lack of Internet access in information access. With a reported cellular mobile density of 70% in 2010, mobile based dialog systems in Urdu have the potential to bypass all three barriers. To create such a system, Urdu speech recognition systems, text to speech synthesis systems and dialog processing systems need to be developed. Currently, dialogue processing system has been successfully developed and now we are developing a speech based service which would provide weather information using mobile phones, aiming to serve rural population of Pakistan, especially the illiterate and print disabled population.

To develop the speech recognition models, we need to collect recordings of various districts of Pakistan from across Pakistan to cover the accent variation. Volunteers are requested to record their responses by contributing a two minute call using their mobile phones.

For this purpose, we have deployed an online recording service, which can be accessed by calling the phone number 042-36821317. The duration of the call is two minutes in which people will be requested to record district names of Pakistan.

Dr. Sarmad Hussain
Professor and Head
Center for Language Engineering (www.cle.org.pk)
Al-Khawarizmi Institute of Computer Science (www.kics.edu.pk)
University of Engineering and Technology (www.uet.edu.pk)
G.T. Road, Lahore
Pakistan
Email: sarmad.hussain@kics.edu.pk