论文部分内容阅读
在当前网络状况下,高效的语音压缩编解码器可节省网络传输带宽,解决网络拥挤问题。通过对语音特性的分析,结合共扼结构代数码本激励线性预测(conjugate structure algebraic code excited linear prediction,CS-ACELP)算法,提出了一种运算量较小但行之有效的话音激活检测算法,设计出了一种基于TMS320VC5409数字信号处理器的语音压缩编解码器。实验结果表明,利用该语音压缩编解码器可将平均比特率降低到约4 kb/s,能很好地满足VoIP中全双工实时语音通信的要求,得到了较好的实际效果。
In the current network conditions, efficient voice compression codecs can save network bandwidth and solve network congestion problems. By analyzing the speech characteristics and combining with the conjugate structure algebraic code excited linear prediction (CS-ACELP) algorithm, this paper proposes a small but effective voice activity detection algorithm, Design a voice compression codec based on TMS320VC5409 digital signal processor. Experimental results show that using the speech compression codec can reduce the average bit rate to about 4 kb / s, which can meet the requirements of full-duplex real-time voice communications in VoIP and achieve good practical results.