Active Voice Detection Using Ridgelet Transform

Abstract

Regularly, voice activity detection is the main crucial parameter inany control system using command of voice speech, spatially inenvironments with noise. The uses voice activity detection method affectsboth, complexity of computational and the overall performance of anycontrol system using command of voice speech. This paper handles theproblem of automatic word boundary detection in both cases (thenoiseless and noisy backgrounds), by proposed voice activity detection ofspeech signals based on using Ridgelet Transform (RT). It uses theInverse Ridgelet Transform (Ridgelet Transform multiply RidgeletTransform), which gives the accuracy in environments with noise. Theexperimental results point up the effectiveness of the used method in lowsignal to noise ratio environments, for command voice speech signaldetection capability.