Improvement of 2.4kbps LPC10 algorithm based on LSF parameters
Xin Zhi Yu, Xingyuan You, Ming Yang, Xiaoling Liu · 2024
Short-wave communication has unstable channel conditions, but it is still widely used in military and diplomatic fields due to its security, high destruction resistance and full coverage characteristics. Therefore, in order to adapt to transmission in shortwave channels, further research on low rate voice coding algorithms under low reliable channel conditions is needed. In this paper, we propose a coding algorithm that converts linear prediction coefficients into line spectral pairs of frequency parameters, and on the basis of the traditional 2.4kbps LPC10 algorithm, we reduce the coding rate to 1.7kbps and score the synthesized results by the PESQ algorithm, and the results show that the improved algorithm yields a voice score of 2.1870, which is an increase of 0.9315 compared to the LCP10 algorithm, with a significant improvement in voice quality. The voice quality is greatly improved.