浅谈语音压缩编码的发展和应用

Talking about the Development and Application of the Phonetic Compressed Encoding Method

苏桃

近30年来，高质量、低速率的语音编码算法不断出现。编码技术种类很多，按照波形编码、参数编码和混合编码的分类方法论述了语音编码的发展进程、各种标准及其应用。[著者文摘]

The phonetic encoding algorithm with high quality and low speed appears continuously in the last thirty years. This paper expounds the developing history, various standards and applications of the phonetic encoding algorithm according to the classification of the wavelike encoding, parameter encoding and compound encoding.[著者文摘]

ACELP语音编码中代数码书的快速搜索算法

A Fast Searching Method of Algebraic Codebook in ACELP Speech Coding

鲍长春 窦庚欣 范睿 刘泽新

为降低ACELP语音编码中代数码书搜索的复杂度,提出了一种基于代数多子码书结构的多路径快速搜索算法.实验结果表明,在不降低编码质量的条件下,这种搜索算法的复杂度仅为全搜索算法的1/128.[著者文摘]

To reduce the the complexity of algebraic codebook searching, a fast multi-path searching algorithm based on algebra sub-codebook is presented. Experimental results show that the complexity of the proposed method is 1 of 128 of full searching method while the quality of speech coder is not affected.[著者文摘]

ITU-T G.723.1语音编码算法分析及优化策略

Analysis and optimization of ITU-T G. 723. 1 speech coder

李纯静 沈保锁

本文介绍了国际电信联盟（ITU）建议中G.723.1低速率语音编码器的基本原理,分析了其中主要模块的实现算法,并对一些计算量较大的模块（基因估计、自适应码本搜索、固定码本搜索）提出优化策略,对编码时间进行优化,使得算法有可能在DSP上实时实现。通过对根据优化策略优化的代码和未优化代码的测试分析,结果显示,优化代码比未优化代码的运行时间减少了21%~31%。[著者文摘]

Introducing the principle of G. 723. 1 low bit-rate speech coder in International Telecom Union （ITU）, analyzing the implementation algorithm of some important modules, and puting forward optimization strategy to these the modules with large computational complexity such as pitch estimation module, the adaptive and fixed code-book research modules, and also optimize the running time of the coder to make it possible to realize the real time implementation of G. 723. 1 on DSP. Experimental results show that the running time of the optimized source code is decreased by 21%-31% in comparison with that of the non-optimized one.[著者文摘]

基于GMM的甚低码率语音编码器

A Very Low Bit-rate Speech Coder Based on GMM

李平 曾毓敏 吴婷婷 吴华玉

提出了一种新颖的基于高斯混合模型（GMM）的甚低码率语音编码系统。该编码器利用GMM对短时语音谱包络进行拟合的方法来对语音进行参数化表示。编码时，语音经预处理、分帧加窗后，再经FFT分析得到分帧语音的信号频谱，并获得平滑谱包络。然后采用GMM对谱包络进行拟合，用GMM参数（均值、方差、权重）对语音谱加以表示。由于GMM参数较少，从而可以使得码率甚低。解码时，根据编码逆运算生成谱包络，浊音信号利用正弦模型加以合成，清音信号经IFFT合成。实验仿真结果表明：该编码器在传输码率降低到2．35kb／s时，仍可获得音质令人满意的解码语音。[著者文摘]

A novel very low bit-rate speech coder based on Gaussian mixture model （GMM）, which is used to parameterize the short-time speech spectrum envelope, is proposed in this paper. In the coding procedure, speech signal is firstly pre-emphasized and segmented. Secondly, the segmented speech is transformed to spectrum domain and the spectrum envelope of the segmented speech is obtained. Then the spectrum envelope is parameterized by GMM. So the segmented speech is represented by the means, covariances and mixture weights of GMM. In the decoding procedure, the spectrum envelope of segmented speech is reconstructed with the inverse method of the coding. Then the speech is synthesized based on the reconstructed spectrum envelope, in which the voiced speech is synthesized by sinusoid model and the unvoiced speech is just synthesized by inverse FFT. Since the segmented speech can be rep- resented by very few parameters of GMM, the bit-rate of the coder is very low. The result of the experiment shows that the proposed speech coder presents a good performance. The quality of the synthesized speech is still satisfying when the bit-rate of the coder is reduced to 2, 35 kb/s.[著者文摘]

G．729语音编码器在DSP上的实时实现

Real- time Implementation of G. 729 Speech Codec Based on DSP

周敬利 赵冕 郭红星

为满足音视频的同时需求，提出了基于TMS320C64X系列DSP的G.729语音编码器软、硬件设计方案，并着重阐述了在实时实现过程中进行优化的关键技术。经测试表明，优化后的单路语音编码占用了5％左右的CPU资源，比优化前降低了80％，保留了更多的资源用于视频编码，为在同一块DSP芯片上实现音频、视频编码提供可能。[著者文摘]

To satisfy the requirement of need for both audio and video simultaneity, a kind of hardware and software scheme of speech encoding based on TMS320C64X DSP is presented. We put our emphasis on key techniques for real - time implementation. The result shows that optimized one - way speechcoder occupies 5% CPU -load ,reduced by 80% compared with the original coder ,reserving more resources for video coding. This is helpful for implementing both speech coder and video coder on one DSP chip.[著者文摘]

新型的神经网络线性预测语音编码算法

A new algorithm for linear prediction coding based on neural network

王涌 何剑春 刘盛

提出一种新型的神经网络线性预测编码算法．针对目前自相关法存在着预测系数解误差以及协方差法存在解不稳定的缺点，算法利用最小均方准则思想显著提高了短时平均误差精度．通过窄带信道将低速率语音编码远距离传输是多媒体语音技术中的重要研究内容，采用语音信号压缩处理是解决低速率传输的有效方法之一，而线性预测编码（LPC）技术是语音压缩参数编码技术的重要内容．从线性预测编码技术入手分析和研究LPC编码技术的原理，阐述了利用最小均方准则思想来提高短时平均误差精度的方法，并介绍了预测系数的自相关求法．最后通过语音合成实验验证了该新型算法既提高了系数解的精度，又保证了系统的稳定jI生．在该算法下预测系数的均方误差比传统的自相关法误差减小20％以上，而且当原始信号频率较高时语音合成的效果更明显，将更加精确地复现原始语音信号．[著者文摘]

A new algorithm of linear prediction coding based on neural network is presented. In contrast with the prediction .coefficients error of self-correlative method and instability of covariance method, this algorithm improves the short-time average precision observably, As the main content of multimedia information compressing technology, long-distance transmission of the speech parameter coding by narrow band signal channel plays an important role. Speech signal compressing is an important section of multimedia information compressing technology. The theory of LPC coding is researched based on the linear prediction coding technology, the least mean square rule is applied for improving the short-time average precision, and the self-correlative calculation of prediction coefficients are introduced. Finally the speech synthesizing experiment is carried out and the result demonstrates that it is not only proved to be an effective way that improves the precision but also ensures systemic stability. Up to 20 percent decrease of the mean square error of prediction coefficient by our method is obtained in contrasting with that of a traditional self-correlation method. When the original signal has a higher frequency, the quality of speech synthesizes with applying this algorithm would be more effective.[著者文摘]

基于低速率语音的自适应联合编码调制方案

Joint source-channel coding and modulation scheme for low-bit-rate speech

彭坦 崔慧娟 唐昆

为了在时变衰落的高误码率无线信道上进行实时可靠语音通信，提出一种基于低速率语音的自适应联合信源信道编码与调制传输方案。该方案根据对合成语音质量影响程度不同自适应地将语音参数划分为不同重要性级别，配合采用不同效率的速率匹配的删余卷积码（rate　compatible　punctured　code，RCPC）进行保护，在正交频分复用（orthogonal　frequency　division　modulation，0FDM）系统的不同子信道上进行传输。各参数均根据当前信道的噪声水平跟踪动态调整，子信道的瞬时噪声根据RCPC译码前后序列符号改变情况进行估计。仿真结果表明：在Nakagami—m时变衰落信道下，该自适应联合编码调制方案可以很好地适应信道地时变衰落特性，在恶劣的信道条件下提供良好的合成语音质量，相对传统分离传输方案，误码率降低20％～30％，语音合成质量明显提高。[著者文摘]

A novel error resilience joint source-channel coding and modulation scheme based on very low bit rate speech coding was proposed to eliminate the transmission performance degradation caused by time-variant fading interference in wireless channels. Based on the different error sensitivity of speech parameters, the output parameters are divided into different level of importance which later will be protected by relevant RCPC （rate compatible punctured code） with diverse efficiency and transmitted through every subchannel of OFDM （orthogonal frequency division modulation） system. The parameters are dynamically selected according to instantaneous channel BER estimated from the bit sequences before and after the RCPC decoding. Simulation results show that, over Nakagami-m fading channels, the proposed scheme is well adapted to time-variant wireless channels, achieving better reconstructed speech quality and BER is decreased at 20%-30%, compared with classic tandem transmission scheme.[著者文摘]

基于DSP的语音编码系统设计

Design of speech coding system based on DSP

郭芙蓉

G.729是国际电信联盟（ITU）于1996年推出的语音压缩标准，G.729A是ITU最新推出的语音编码标准G.729的简化版本。根据TMS320C5410的特点，提出了采用G.729A语音编解码算法设计的语音压缩系统。给出了系统的硬件结构，包括手动复位电路、晶振电路、电源电路的设计，以及CPLD在DSP系统中的控制分析。此外，还给出软件编程，以及定点优化时的一些原则和方法。[著者文摘]

G.729A is a standard of speech coding produced by the international telecom union （ITU） in 1996. G.729A is the new speech coding standard, which is the simplified edition of G.729. Based on the architecture of TMS320C5410, this paper designs a speech coding system using G. 729A decode algorithm. The hardware structure is given, which includes the design of detailed manual reset circuit system, crystal oscillation circuit, power circuit, and the control of CPLD in DSP data acquisition system. Moreover, the paper gives out software programming and some princeples and ways on how to implement it in the fixed point program.[著者文摘]

CVSD编码的语音质量受跳频速率影响研究

Research on Effect of FH Rate on Speech Signal＇s Quality Based on CVSD Coding

黄小刚[1] 曾现巍[2] 陈建忠[2]

针对采用CVSD编码的数字跳频系统的抗干扰能力受跳频速率的影响进行了研究。给出了CVSD编码的基本原理以及慢跳频系统数据传输模型;提出了语音质量衡量准则和跳频系统等效基带仿真方法,同时指出了CVSD编码参数的优化选取方式;通过仿真具体分析了跳频速率对数字跳频系统抗部分频带噪声干扰能力的影响,并得出了结论。[著者文摘]

This paper has researched the effect of hop-rate in the FH anti-jamming system based on CVSD coding.It presents the principle of CVSD coding and the model of SFH,suggests the rule for measuring speech signal＇s quality and the method of how to simulate the FH system with baseband model,and points out the optimized parameters selecting of CVSD coding.In the end,the paper analyses the performance of FH system with different Frequency-Hopping rates on the part-band-noise jamming channel,and makes a conclusion.[著者文摘]

ACELP语音编码器中增益码书的建立方法

Method for Establishing the Gain Codebook in ACELP Speech Coder

鲍长春 窦庚欣 范睿 朱恒

为提高ACELP语音编码器中激励增益的量化性能,基于广泛应用的代数码激励线性预测语音编码模型,提出了一种具有一般意义的激励增益码书的建立方法,该方法可应用于各种不同速率ACELP语音编码器中.实验结果表明,使用该方法建立的6 b增益矢量码书与8 b标量增益码书的性能相当.[著者文摘]

In order to improve the quantization quality of the excitation gains in ACELP speech coder, we present a general method for establishing the gains codebook in widely used ACELP speech coders. This method can be applied to ACELP speech coder at any bit rate. Experimental results show that the performance of 6 bits gains vector codebook, which is established by the purposed method, is equal to that of 8 bits gains scalar codebook.[著者文摘]