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{203348,
        author = {Chinmay Joshi and Kapil Joshi and Ojas Joshi and Aditya Dhule and Supriya Telsang},
        title = {Retrofitting Legacy Homes with AI-based Offline Voice Control},
        journal = {International Journal of Innovative Research in Technology},
        year = {2026},
        volume = {12},
        number = {12},
        pages = {11910-11915},
        issn = {2349-6002},
        url = {https://ijirt.org/article?manuscript=203348},
        abstract = {Numerous current smart-home systems rely on proprietary ecosystems and cloud-based voice assistants, which necessitate constant internet connectivity and frequently require replacing outdated appliances or electrical infrastructure. These restrictions raise issues with latency, dependability, and user privacy and make it challenging to implement such systems in older homes. A retrofit-focused offline voice-controlled home automation system that can function without internet access or specific smart appliances is presented in this paper. The suggested system makes use of an AI speech recognition module that can recognize commands and identify keywords on-device, allowing for direct voice control of household appliances using pre-programmed commands. Relay-based control interfaces are used in the architecture's integration with current electrical switchboards to automate traditional appliances without the need for centralized hubs or rewiring. Four household loads, such as lighting and fan circuits, were controlled by a prototype implementation. An average command recognition accuracy of roughly 90% with a response latency of 100–150 MS was attained in an experimental evaluation conducted in a typical indoor setting. While recognition accuracy declined for far-field voice input beyond about 10 m in noisy environments, performance remained consistent under typical household indoor conditions. The findings show that offline voice-controlled retrofitting of legacy home infrastructure is feasible.},
        keywords = {retrofit, smart homes, offline, voice recognition, IoT, automation},
        month = {May},
        }