摘要
研究了小波包变换在声音编码中的应用，通过利用小波包变换在时域和频域都具有良好局部性的特点，提出了一种新的中低码率混合域声音编码算法。这是一种可调码率算法，它根据输入信号段的特性来进行变换和编码，使得编码过程可以同时在时域、频域以及从时域向频域过渡的任一时频域中进行，从而使编码算法不仅在量化编码阶段，而且在变换阶段就对输入信号具有自适应性。其编码效果比在单一固定时频域中进行变换与编码有较大改善，有助于和高码率的声音编码算法相接轨。
The use of wavelet packet transforms in audio coding is studied, and a new medium and low bit rate mixed domain audio coding algorithm has been given based on the good localization of wavelet packet transforms in both time domain and frequency domain. It is a rate adaptive algorithm, and the transform and the coding are done according to the characteristics of the input signal segments, which results in an adaptive coding in time domain, frequency domain or any one of the intermediate time frequency domains between them. Therefore, the algorithm is adaptive to input signals not only in quantization and coding, but also in transform. The performance of the proposed algorithm is improved compared with the one in single time frequency domain, and it is helpful to be combined with high bit rate audio coding algorithms.
出处
《清华大学学报（自然科学版）》
EI
CAS
CSCD
北大核心
1998年第9期28-31,共4页
Journal of Tsinghua University(Science and Technology)
基金
国家自然科学基金
关键词
声音编码
小波包变换
混合域
audio coding
wavelet packet transform
time frequency domain
bit rate
bit allocation
