Wideband speech coding based on mixed excitation linear prediction(MELP)

• Main Author: Lim, Weiran.
• Other Authors: Koh, Soo Ngee
• Format: Theses and Dissertations
• Published: 2008
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing
• Online Access: http://hdl.handle.net/10356/4712
• Institution: Nanyang Technological University

The purpose of this research has been to study the feasibility and effectiveness of using the Mixed Excitation Linear Prediction (MELP) model for coding wideband (7 kHz) speech signals at the bit rate of 8 kbps. In order to achieve a good subjective quality for the decoded speech while maintaining a low operating bit rate at the same time, modifications to the pitch estimation, voiced/unvoiced strength analysis, linear predictive (LP) analysis/synthesis of the original MELP model are proposed and examined in detail. The quantization of the parameters of the modified model is also studied. Based on the proposed modifications, an 8.4-kbps and an 8-kbps wideband MELP coders are developed. The quality of the decoded speech obtained from our proposed wideband coder is assessed in terms of subjective as well as objective measures.