Bridging the digital divide: a survey on development of spoken language technology, a northeast India perspective

Abstract The modern advancement of language technology now drives the rapid development of native languages throughout Northeast India. Indigenous northeastern languages of India encounter major obstacles in speech processing because of their diverse linguistic structure as well as complex phonetics and insufficient resources. The research examines the existing developments regarding speech technologies and methodologies which operate in automatic speech recognition (ASR) and natural language processing (NLP) systems for Northeast indigenous languages. The research investigates different approaches for linguistic resource development together with different corpus types used and diverse data collection methods to create ASR systems for languages with minimal resources. It also highlights new machine learning approaches used for feature extraction and modelling techniques, for the development of automatic speech recognition technology (ASR). This study fosters the development of the spoken language technology research, which helps preserve languages, empower indigenous groups in the quickly changing digital context, and promote digital inclusion. As a result, we believe that our study serves as a foundation for researchers who are interested in ASR research.