Application of Artificial Intelligence AI Recognition for Language Maintenance in HO Language

Abstract

newlineABSTRACT newlineThe world of language technology is rapidly changing, but the benefits of language technology in newlinethe world are highly uneven. Though willing to adopt these innovations, indigenous language newlinecommunities are left on the periphery because of the dire lack of data and because standard newlineorthographies and scarce computational resources are unavailable. Linguistic preservation and newlinecultural survival have emerged as a matter of urgent concern due to the proliferation of indigenous newlinelanguage technologies. This is particularly acute in India, where tribal languages like Ho is being newlinerapidly eroded, posing a threat not only to linguistic diversity, but also to the continuity of cultural newlineknowledge systems stored within them. It is against this backdrop that this thesis examines how newlineArtificial Intelligence (AI) can help document, revitalize and sustainably use these underresourced newlinelanguages, placing the technology as a cultural partner in opposition to the idea of a newlinetechnological neutrality. Maintaining native languages is a cultural requirement as well as a newlinetechnological challenge of the digital world. Ho is one of the major languages of the Munda branch newlineof the Austroasiatic family that is underrepresented in computational research. Their problems newlineinclude broken writing systems, sluggish generational transmission and almost no robust digital newlineresources. To overcome these challenges, this work establishes a multimodal AI framework, newlineinformed by interdisciplinary scholarship integrating linguistic knowledge, the latest deep learning newlinemodels and participatory field practice with tribal peoples.