hhspectrum

本资料是自己编写的基于MATLAB语音工具箱的DTW动态时间规整的说话人识别，具有很高的识别率(This information is written in their own voice-based MATLAB toolbox DTW dynamic time warping speaker recognition with high recognition rate)

mfcc.m 用于语音识别上的mfcc特征参数提取函数。(mfcc.m mfcc characteristic parameters for speech recognition on the extraction function.)

程序实现的功能是：一种改进型的LMS算法！具有收敛速度快的特点(Procedures for the realization of the functions are: an improved algorithm of LMS! With the characteristics of fast convergence)

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.)

在提供音频的情况下，利用双音多频确定拨号音。(In the provision of audio cases, the use of dual-tone multi-frequency determine the dial tone.)

一个简易的端点检测程序，已经实现，包括基于过零率、短时能量的双门限判决方法。(a simple endpoint detection procedures has been achieved, including on the basis of zero rate Short-term dual energy threshold judgment method.)

对基于残留噪声去除的谱减法的编程，实现了高去噪水平(The program of spectral subtraction based on residual noise removal has achieved high denoising level.)

AMR-WB+编解码源代码，内含工程文件，可直接运行(AMR-WB+ codec source code, containing the project files can be directly run)

微软研究院的说话人识别工具包，包括GMM-UBM、I-Vector。其中demo_gmm_ubm_artificial.m和demo_ivector_plda_artificial.m为生成模拟特征参数进行训练与识别的教学示例，十分适合初学者学习说话人识别基础算法。具体使用方法请看内部文档。(Microsoft Research s speaker recognition toolkit, including GMM-UBM, I-Vector. Demo_gmm_ubm_artificial.m and demo_ivector_plda_artificial.m which generates an analog characteristic parameters for example teaching training and recognition, very suitable for beginners to learn the basic algorithm for speaker recognition. See the specific use of internal documents.)

做毕设是用到的gmm的matlab程序，测试过的 ，绝对可以用(Complete set is used to do the matlab program gmm, tested, can surely be)