Speech enhancement using wavelet packets

• Main Author: Tan, Kwang Jong.
• Other Authors: Soon, Ing Yann
• Format: Theses and Dissertations
• Language: English
• Published: 2008
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Electronic systems::Signal processing; DRNTU::Engineering::Electrical and electronic engineering::Computer hardware, software and systems
• Online Access: http://hdl.handle.net/10356/13238
• Institution: Nanyang Technological University

A new speech enhancement algorithm for speech corrupted by background noise based on Wiener filtering and discrete wavelet transform was developed. Wavelet Packet Analysis Tool of MATLAB was first used to break down the noisy speech into different sub-band components. Next, the sub-band components of the noisy speech were passed to Wiener smoothing filter. Finally, the filtered components were reconstructed back to time-domain to obtain the speech signal estimate. Selection of the different sub-bands was based on existing information available in the method used for Articulation Index calculation. The proposed algorithm is to be appeared in the Singapore ICCS '98 Nov issue [1].