Literary and colloquial tamil speech identification for inclusive human computer interaction

Abstract

The evolution of any culture in the world is bound to bring about the evolution of its language. This fact is also true in the case of Tamil, an Indian language wherein the ancient literary form evolved into its modern colloquial form. Literary Tamil (LT), which has a rich legacy of grammar and literature is considered archaic, yet it is currently used as a formal language in government orders, state assemblies, news, lectures, literary works, and sometimes in the media. Colloquial Tamil (CT), which is a derivative of literary Tamil, is not very popular in the written form, yet it is ingrained in the day-to-day vocal communication of the common man. The choice of picking one of either literary or colloquial Tamil for human computer interaction (HCI) systems is critical. If only literary Tamil is addressed, the comfort of the common man, and the inability of illiterates are ignored. On the other hand, if only colloquial Tamil is addressed, it will fail to support the numerous movements that strive to preserve the prestigious status of literary Tamil. Yet, if both preservation and inclusivity are to be supported, the only solution is to address both literary and colloquial Tamil. newline