Speech Processing Technology: A Comparative Study of Pakistan and India.

This paper focuses at Natural Language Processing technology research in Pakistan and India with specific focus on research groups working on Speech Processing Technology and initial efforts to enrich the linguistic database of regional languages in research institutions, universities and multinational organizations. Speech Processing Technology is a specialized research field and is increasingly becoming popular in areas of information communication technology, humanoid robots, artificial intelligence, and consumer services in banking, telecom sector, industries, health (speech assistance for the disabled) and education. It comprises Speech Comprehension Technology and Speech Generation Technology. For both these technologies to perform efficiently, reliable linguistic database is essential in targeted languages. This study highlights the potential projects of research groups and the strengths and weaknesses in this research area in Pakistan and India, and how both countries can benefit from each other's experience in speech processing research. There are striking similarities in the speech utterances of languages spoken in Pakistan and India like Sindhi, Kashmiri, Urdu, Hindi and Punjabi. In Pakistan, Natural Language Processing and Speech Processing Technology is at a very early stage as compared to India, which has initiated research projects in Sanskrit, Hindi and more than twenty regional languages. In both countries, there is a strong need to develop a large linguistic database of these languages before any further research in processing systems for speech comprehension or speech generation can be initiated while highlighting the issues related with lack of sustained research projects in Pakistan. This paper suggests areas of improvement in Pakistan by enriching the linguistic database in the indigenous Pakistani languages for market based research for improvement in consumer services in Speech Processing Technology in Pakistan. [ABSTRACT FROM AUTHOR]