LMS_PI

基于峭度的盲分离开关算法的matlab源代码程序(Blind Source Separation Based on Kurtosis switch algorithm )

压缩包中包括3个用于对语音增强中，增强语音质量测试的程序：分段信噪比提高测量、对数谱距离测量、信噪比测量。具体说明，请看文件文件说明。(Compression package, including three for speech enhancement, the enhanced voice quality testing procedures: Sub-noise ratio improve measurement of the distance spectrum measurement, signal to noise ratio measurement. Specify, see the paper documentation.)

基于倒谱短时部分反映了语音的声道特性，先用汉明窗取一帧语音，然后经变换得到语音倒谱，将倒谱短时部分取出，进行正交反变换后将得到声道的对数谱，即得到语音频谱的包络。将频谱包络和频谱画在一张图上，有很好的对比效果。获取的包络效果十分好。(Based Cepstral partly reflects the short channel characteristics of the speech, first take a Hamming window with a frame of speech, and speech cepstrum obtained by converting the cepstrum short segment out inverse orthogonal transform to obtain channel will of the spectrum, i.e. to obtain the envelope of the speech spectrum. The spectral envelope and spectral painted on a chart, there is a good contrast. Get the envelope effect is very good.)

VAD去噪算法，可以有效的过滤静音和无效的语音，同时消除了一些背景噪声。(VAD denoising algorithm, can effectively filter invalid mute and voice, while eliminating some background noise.)

采用广义旁瓣抵消（GSC）自适应波束形成方法实现时域和频域滤波，采用LMS自适应算法，最终实现语音增强。（文件中包含纯净语音及不同信噪比的带噪语音）(Generalized sidelobe canceller (GSC) adaptive beamforming method to achieve time-domain filtering using the LMS adaptive algorithm, and ultimately the speech enhancement)

这 里主要对LMS算法及一些改进的LMS算法（NLMS算法、变步长LMS算法、变换域LMS算法）之间的不同点进行了比较，，在传统的LMS算法的基础上发 展了LMS算法的应用。另一方面又从RLS算法的分析析中对其与LMS算法的不同特性进行了比较。 (Here the main difference between the LMS algorithm and improved LMS algorithm (NLMS algorithm, variable step size LMS algorithm, the transform domain LMS algorithm) comparison, the traditional LMS algorithm based on the development of the application of the LMS algorithm . On the other hand and from its different characteristics of the LMS algorithm of the analytical analysis of the RLS algorithm.)

基于vc++开发，dtw语音识别,MFCC参数(DTW recognice based on vc++)

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.)

Use of repeater transponder controls, WebForm controls

基于DTW算法实现的语音识别程序，可以识别数字0 9，以及“正弦波，方波，三角波”，并将波形识别结果装入TXT文件中！(DTW algorithm based on speech recognition program that can recognize digits 0 to 9, and sine, square, triangle, and waveform recognition result is loaded into a TXT file!)