A focus on codemixing and codeswitching in Tamil speech to text

Abstract

This paper attempts to develop an application that converts Tamil language speech to Tamil text, with a view to encourage usage and indirectly ensure linguistic preservation of a classical language. The application converts spoken Tamil to text without auto-correction, code-mixing or code-switching. Tamil is a syllabic language, similar to other Indian languages and some unique features such as instances of allophones, short & long vowels and lack of aspirated stops produces some challenges in developing a speech to text app. This project is a technology demonstration of a complete web application, which, when perfected, could be used to act as a teaching tool to encourage correct pronunciation of syllables and words for native and non-native Tamil speakers. A report by the Business Standard India e-publication in the year 2019 highlighted the decline in the usage of the Tamil language, and indeed separate reports in Singapore and Malaysia, which both have large numbers of Tamil speakers indicate that there have been concerns about the relevance and usage of the Tamil language as a spoken means of communication among the community. The research maintains that it is important to maintain the utilization of Tamil language via technology to help preservation of one of the oldest surviving languages in the world. The work further emphasizes on the indigenous design considerations for such applications which may be different to traditional software engineering approaches..