Analysis of Stop Consonants in Indian Languages Using Excitation Source Information in Speech Signal

In this paper we propose excitation-based features for extr acting information about the manner of articulation for stop co ns nants. The excitation-based features are derived from very low frequency information in the signal and also from the normal ized error computed from the linear prediction residual. Th e proposed zero-frequency filtered signal brings out the regi on of glottal activity during excitation. Likewise, the norma lized error helps to distinguish regions of noise and pure voicing . These nonspectral methods of analysis of stop consonants se em to provide additional and some better features over the feat ur s derived from the traditional methods based on short-time sp ectrum analysis.