Overview

• Researchers at Beijing Institute of Technology have developed a low-cost visual microphone that detects sound-induced vibrations on everyday objects by analyzing reflected light patterns.
• The system uses Fourier-based localization and computational imaging to track minute surface motions and applies signal-processing filters to correct low-frequency distortion.
• Laboratory tests demonstrated clear reconstruction of Chinese and English speech and a segment of Beethoven’s Für Elise from paper cards and leaves placed half a meter from the source.
• With a data rate of just 4 MB/s, the device minimizes storage demands and supports long-duration recording without bulky cameras or lasers.
• Ongoing work aims to enhance sensitivity, extend detection range and shrink the setup for real-world applications such as through-barrier communication and medical monitoring.