Sahar Khaksar Haghani - Amirkabir University of Technology
Seyed Mohammad Ahadi

This paper presents a study of noise-robust voice activity detection (VAD) utilizing combination of feature vectors extracted from speech signals. Conventional VADs are sensitiveto non-stationary noise especially in low SNRs. Also situations such as cutting off of unvoiced regions of speech and randomoscillation of VAD decisions are unavoidable. To overcome these problems, the proposed algorithm utilizes measures such as energy differences, periodicity, zero crossing rate, and spectral differences between different sound frames. The performance of the proposed VAD algorithm is tested on realspeech signals. Comparisons confirm that the proposed VAD algorithm outperforms the conventional VADs, especially in the presence of background noise

Voice activity detection, feature combination,robustness;