Virtual Booth

The number of individuals with disabilities in Malaysia is steadily increasing, with many facing daily challenges in their homes. Voice-controlled home devices offer significant assistance, yet most require English input, posing a barrier for non-English speakers. This study introduces a machine learning-based home assistant that operates in 'Bahasa Melayu,' accommodating dialects such as Kelantan, Perak, Penang, and the standard dialect. Data was collected via smartphone, and speech recognition was performed using Artificial Neural Networks on the Edge Impulse platform. Deployed on Arduino BLE 33 Sense, this system controls home devices like lights and fans, achieving high accuracy in ANN classification and real-world application. This innovation promises to enhance the quality of life for disabled individuals in Malaysia, promoting greater independence and ease of living.