Copyright © 2026 Authors retain the copyright of this article. This article is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

@article{197728,
        author = {M Janani and AJ Somesh Kumar and G Vivek Gandhan and A Sanjai and M Vishwa},
        title = {Voice first AI vernacular assistant for goverment schemes and entitlements},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {11},
        pages = {10085-10088},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=197728},
        abstract = {Accessing government schemes is often hindered by digital vernacular speech to connect citizens with relevant government schemes. By prioritizing local languages and robust error-literacy gaps, language barriers, and complex user interfaces. This paper presents a Voice-First AI Vernacular Assistant designed to democratize access to government entitlements. Developed as a Single Page Application (SPA) using React 19 and Vite, the system leverages the Web Speech API and a custom Natural Language Processing (NLP) engine to process multi-lingual speech. Key innovations include regular expression-based speech disfluency cleaning for English, Hindi, and Tamil, alongside dynamic phonetic language detection. A Profile Extraction Engine derives structured user data from unstructured speech, fueling a precise, point-based scoring algorithm. To ensure reliability, strict confidence guardrails prevent AI hallucination. The resulting platform offers a highly accessible, intuitive, and accurate solution for seamless scheme discovery.},
        keywords = {Voice Assistant, Vernacular AI, E-Governance, Natural Language Processing, Zero-Hallucination, React 19, Speech Disfluency Cleaning.},
        month = {April},
        }