The Enhanced Speech Recognition in Automated Home Lighting System using Adaptive Time-Frequency Domain Noise Removal Algorithm Filter

Citation

Siddik, Sk Abu Baker and Wan Husin, Wan Nor Al-Ashekin and Muniandy, Thagirarani (2024) The Enhanced Speech Recognition in Automated Home Lighting System using Adaptive Time-Frequency Domain Noise Removal Algorithm Filter. International Journal on Robotics, Automation and Sciences, 6 (1). pp. 20-28. ISSN 2682-860X

[img] Text
548 - Published Version
Restricted to Repository staff only

Download (2kB)

Abstract

Numerous studies have explored speech recognition performance in Smart Home environments. However speech recognition accuracy diminishes when voice commands are captured in noisy areas of the home. This study aims to enhance speech recognition performance in such noisy environments. Instead of relying on remote control signals, a Bluetooth system is employed for short-range wireless communication to identify speech commands. Various sound levels are measured in decibels (dB) at different distances using the Smart Noise Application. A filter algorithm with Adaptive Filtering is used to minimize unwanted noise. The algorithm uses Adaptive Time-Frequency Domain Noise Removal (TFDNR) to mitigate background noise. Overall, the integrated system comprising Smartphone, Bluetooth, Arduino microcontroller, and noise detection software exhibits improved performance compared to previous studies, highlighting its potential for seamless smart home automation.

Item Type: Article
Uncontrolled Keywords: Speech Recognition, Bluetooth, Decibels, Noise, Adaptive Filter
Subjects: T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK1001-1841 Production of electric energy or power. Powerplants. Central stations
Divisions: Others
Depositing User: Mr. MUHAMMAD AZRUL MOSRI
Date Deposited: 06 Sep 2024 00:30
Last Modified: 06 Sep 2024 00:30
URII: http://shdl.mmu.edu.my/id/eprint/12964

Downloads

Downloads per month over past year

View ItemEdit (login required)