Enhanced wavelet based audio coder
Ahmed H. Tewfik, Murtaza Ali · 2002
Several enhancements on the wavelet based audio coder proposed by Sinha and Tewfik (see. Proceedings of the Int. Conf. of Acoust., Speech and Signal Proc. Society (ICASSP), vol.1, p.197-200, 1993) are discussed. These enhancements include use of temporal information in masking for audio segments with sudden change of amplitudes in them, efficient quantization of the variances and run-length coding for the zeros. Experiments suggest that these enhancements give almost transparent compression of monophonic CD quality audio signals (sampled at 44.1 kHz) at bit rates of 55-63 kbits/sec. We also discuss a real time implementation of a simplified version of the wavelet based audio coder at somewhat higher bit rates (64-78 kbits/sec).>