Design and Implementation of a Hybrid Voice- and IoT-Based Smart Home Control System Using Local Speech Processing to Reduce Cloud Dependency

Abstract

Electrical devices are an important part of modern life. However, forgetting to turn off electrical appliances is still one of the causes of house fires in many areas. This research aims to determine the performance and design specifications of a home electrical appliance control system using a voice recognition sensor based on the Internet of Things (IoT). The research method used is an engineering method with several steps, including problem identification, conceptual design, system design, detailed design, prototype building, and system testing. The system is built using an Arduino Uno microcontroller, ESP32 module, voice recognition sensor, and relay module, which are connected to the Blynk application as a user interface for real-time communication. The system has two modes: voice control mode and manual control mode. In voice control mode, devices work based on voice commands from the user. In manual mode, users can control and monitor devices remotely through the Blynk application using a switch. The test results show that the system achieved an accuracy of 90% and demonstrated good consistency under controlled testing conditions. Furthermore, the system operates effectively and has demonstrated a high level of functional success during testing. These results show that the IoT-based control system using a voice recognition sensor works well and is reliable for controlling home electrical appliances.