一种改善激励源的1.2kb/s语音编码算法及其实时实现

An Improved 1.2 kb/s Speech Coding Algorithm and its Real-time Implementation

为了适应无线通信等低速语音编码场合 ,吸收 2 .4kb/s MELP算法优点 ,提出了一种改善激励源的 1 .2 kb/s混合激励线性预测算法 ,该算法在模型结构、LSP量化及解码等方面较传统的 2 .4kb/s LPC算法有了较大改进。同时 ,选用 TMS3 2 0 VC5 4 1 6 DSP芯片实时实现了以该算法为核心的声码器。客观音质测试和非正式主观测试结果表明 ,该算法在自然度、清晰度和抗噪声等方面明显优于 2 .4kb/s LPC算法 ,是低速率下一种良好的编码方案。

In order to meet the needs of wireless communication, a new 1.2 kb/s speech coding algorithm based on mixed excitation linear prediction is proposed in this paper. Compared with the traditional 2.4 kb/s LPC algorithm, this new algorithm improves several aspects such as model structue, LSP quantization and decoding. This algorithm has been implemented on TMS320VC5416 DSP processor. Objective speech quality testing and informal listening testing results show that the intelligibility, naturalness and anti-noise performance of the 1.2 kb/s speech coding algorithm excel 2.4 kb/s LPC algorithm.