Research Project Title:
Implementation of a Voice Activity Detector for Energy Equalization
abstract:Energy Equalization (EEQ) is a signal-processing scheme that combats the reduced dynamic range of hearing-impaired listeners. It reduces amplitude fluctuations of noisy speech without introducing excessive distortion by normalizing the fluctuating short-term signal energy to be equal to the long-term average signal energy, thus amplifying low levels of speech. Our research focuses on implementing a Voice Activity Detector (VAD), which will classify signals as speech or noise prior to EEQ processing, with the goal of implementing EEQ only when speech is present.We will incorporate MATLAB’s VAD G729 algorithm into the current algorithm and compare EEQ processed signals with and without VAD. We will conduct acoustic analyses of both types of processed signals to evaluate the effect of VAD.
“Through this SuperUROP project, I want to gain more experience in signal processing and contribute to a field I am interested in. I have taken signal processing courses and I want to expand on that knowledge with real-world applications. I hope to publish a paper by the end of the SuperUROP and become a better researcher.”