摘要
为了将低比特率语音编码算法应用于实际的语音处理系统,开发了一种基于TMS320VC5410定点DSP 的实时低速率语音压缩系统.语音编码方法采用一种高质量的4 kb/s散布脉冲CELP算法.用汇编语言编写的定点DSP程序能在目标平台上实时完成语音数据的采集、处理和回放功能.整个系统大约占用31 K存储空间, 算法复杂度达到指令数为5.42×107条/s.客观质量测试结果显示,DSP定点程序处理后的语音质量非常接近于浮点C语言处理后的结果,表明该语音处理系统能满足实时语音压缩的需求.
In order to apply the low bit-rate speech coding algorithm to the real speech processing system, a real-time speech compression system based on TMS320VC5410 fixed-point DSP at low bit-rate is introduced in this paper. A high-quality 4kb/s dispersed-pulse CELP algorithm is adopted in this system. The functions of data collection, processing and playback are implemented by fixed-point DSP programs in assembly language. About 31K memory spaces are consumed by this system and the complexity of this algorithm is about 54.2 million instruments per second. Objective test indicated there is no perceptual difference for the reconstructed speech quality compared with the original C programming result. So this speech processing system can be applied to real-time speech compression.
出处
《北京工业大学学报》
EI
CAS
CSCD
北大核心
2006年第5期385-390,共6页
Journal of Beijing University of Technology
基金
北京市教委科技发展项目(KM200310005024)
国家自然科学基金(60372063)
北京市自然科学基金(4042009)
关键词
语音编码
码激励
线性预测
数字信号
speech coding
code excited
linear prediction
digital signal