Text-to-Speech Generation with Multilingual Assistance and AI Integration for Information Generation
Authors: Dr. JayaSudha K, Appu Gowda B S, Meghana B R, Geetha U, Ashritha G N
DOI: 10.87349/ahuri/181031
Page No: 1-12
Abstract
Multilingual speech technologies are critical to enabling seamless communication across languages, empowering users with diverse linguistic backgrounds, and supporting accessibility in digital systems. This paper presents a complete multilingual Text-to-Speech (TTS) architecture enhanced with Automatic Speech Recognition (ASR), Neural Machine Translation (NMT), and AI-driven information extraction. Unlike conventional TTS systems that simply convert text into speech, this integrated pipeline analyzes speech, extracts meaningful information, translates content to a target language, and finally generates expressive and natural audio. We describe each subsystem in detail, the engineering decisions behind the composite pipeline, and the challenges encountered in multilingual deployment settings.
