Pitch Modification of Speech Signal Using Source Filter Model by Linear Prediction for Prosodic Transformations

Abstract

This paper proposes a pitch modification technique based on the use of source filter model (SFM) by Linear Prediction. A complete solution for pitch modification is described, included estimation of model parameters, pitch detection, unvoiced/voiced (U/V) classification and speech synthesis at several pitch values. An automatic U/V classification based on the use of the zero crossing rate (ZCR) and the Energy computation is proposed. The satisfactory performance of this technique is evaluated by listening tests using sentences in Arabic Language. The limit range of pitch modification for natural synthesizing is found.