Wavelet decomposition of voiced speech and mathematical morphology analysis for glottal closure instants detection

This paper presents a robust algorithm for glottal closure instants (GCIs) detection of speech signals. The algorithm uses a multi-scale analysis based on a dyadic wavelet filterbank. Significant minima and maxima of the filtered signals are localized at each scale using adaptive mathematical morphology transformation of erosion. With reference to the GCIs detected from the laryngograph signal, a robust strategy for GCI localization was deduced. Each GCI is determined as the position of a minimum suitably chosen on one of the outputs of the different filters. This choice aims to ensure the best accuracy and reliability even for weak glottal effort.