voicebox

本程序介绍了一种改进的LMS算法-均一化的LMS算法，与传统的LMS算法比较收敛性更好，并由程序给出了比较图(This procedure introduced an improved LMS algorithm- homogenization of the LMS algorithm, with the traditional LMS algorithm convergence better, by the procedures given comparison)

三种降噪，回声和变调的语音信号处理，含UI界面(Three kinds of noise reduction, echo and tone of voice signal processing, including UI interface)

GMM的说话人识别系统，识别效果90 多(GMM speaker recognition system, the recognition effect more than 90 )

1、文件夹中包含了经典功率谱估计和自适应均衡算法两个实验的所有程序。 2、R.m、LMS.m、LMSmain.m为自适应均衡算法的程序： R.m用来计算输入信号的自相关矩阵及其特征值； LMS.m为时域LMS算法，用统计的方法仿真得出不同信道参数和不同步长下的学习曲线； LMSmain.m为实验主程序，按照实验要求中的具体数据得到实验结果和曲线。 3、functionx.m、fzhouqitu.m、spectrum.m、bt.m、bart_lett.m、welch.m、SPECTRUMmain.m为经典谱估计的程序： functionx.m产生需要进行谱估计的函数； fzhouqitu.m用来计算信号周期图的函数； spectrum.m是用周期图法进行谱估计的函数 bt.m是用BT图法进行谱估计的函数 bart_lett.m是用BARTLETT法进行谱估计的函数 welch.m是用WELCH法进行谱估计的函数 SPECTRUMmain.m是主程序，按照实验要求中的具体数据得到实验结果和曲线。 (a document folder contains the classic power spectrum estimation and adaptive equalization algorithm for the two experiments all the procedures. 2, R.m, LMS.m, LMSmain.m adaptive equalization algorithm for the procedure : R.m used to calculate the input signal and the correlation matrix eigenvalue; LMS.m too LMS algorithm, use statistical simulation method come to a different channel parameters and the synchronous learning curve; LMSmain.m main program for the experiment, according to the experimental requirements of the specific data and the curve of the experimental results. 3, functionx.m, fzhouqitu.m, spectrum.m. bt.m, bart_lett.m, welch.m. SPECTRUMmain.m classical spectrum estimation procedures : functionx.m have a need for spectrum estimation function; fzhouqitu.m used to calculate t)

回声消除算法比较，仿射投影APA与NLMS算法比较(Echo cancellation apa nlms)

完整的基音识别程序,经过修改,测试正确。包含基音检测，DTW，LPCC，MFCC提取等(integrity of the pitch identification procedures, the result of the revision, the test correctly. Pitch Detection included, DTW, LPCC Features, MFCC extraction)

近场的MUSIC算法，可用于近场噪声源识别(Near-field MUSIC algorithm can be used for near-field noise source identification)

源－目标说话人声音转换的学位论文：源－目标说话人声音转换则是指将源说话人的语音模式转换成某个指定的特定目标说话人的语音模式，即保持源说话人原有的语音信息内容不变，使转换后的语音具有目标说话人的声音特点。(Source- target speaker voice conversion refers to the source speaker voice patterns into a designated target speaker-specific speech patterns, that is, the source speaker to keep the original voice message unchanged, so that the converted speech with the goal of the speaker s voice characteristics.)

应用LBG算法对语音信号进行矢量量化。本压缩包共有两个主文件，training.m对训练数据进行处理并得到四个初始码本，quantizing.m对待量化数据进行矢量量化。其余为自编功能函数。(application of voice signal VQ. The compressed into two main documents, training.m right training data to be processed and four of the initial code, quantizing.m treat quantitative data for vector quantization. The remainder self-function.)

speech file to be analyzed Format : binary file 16 bit-samples 240 samples per frame (speech file to be analyzed Format : 16 bit binary file-samples of 240 samples per fram e)