摘要
目前2.4kbps的混合激励线性预测(MELP)语音编码方法已经被确定为美国新的联邦语音编码标准。本文 提出了一种改进的MELP语音编码方法,利用滤波器相似度和基于LPC系数分类的矢量量化技术,可以把MELP的 码率降到1.7kbps以下,仍有较好的合成语音质量。
Mixed Excitation Linear Prediction(MELP) coder has been adopted as the new US Federal Standard at 2.4kbps. An improved MELP speech coder is proposed in this paper. The concept of similarity between linear filters and a new method of vector quantization based on classifying the LPC coefficients are proposed. By these two methods the bit rates of MELP coder can be reduced to as low as 1.7kbps without evident damage to the synthetic speech quality.
出处
《电路与系统学报》
CSCD
2003年第1期101-104,共4页
Journal of Circuits and Systems
基金
国家自然科学基金资助项目(60272039)
安徽省自然科学基金项目(901042205)
