speech-signal-processing

实现多个人的说话人识别，基于gmm模型，用mfcc参数训练(voice conversion based on gmm)

基于麦克风阵列的声源定位一维music算法源代码(Based on the microphone array beamformer one-dimensional music algorithm source code)

matlab实现的基于小波变化的语音去噪源码，可以去除语音中的背景噪声。(Matlab based on the wavelet changes in the voice de-noising source, you can remove the background noise in the voice.)

实验示例是基于语音中的mfcc，语音倒谱特征来进行聚类，先利用训练样本来计算训练样本聚类中心（用到了lbg算法），之后再进行分类。 注意：使用代码时需要自己更改文件路径。(This example is based on the MFCC in speech and the feature of speech Cepstrum to cluster. First, the training sample is used to calculate the training sample clustering center (using the LBG algorithm), then the classification is then carried out.)

voice recognition using mfcc

语音库开发工具源码，例程程序创建机读SAPI.SpVoice对象开发语音库。(Speech development tools source , routine program to create machine-readable SAPI.SpVoice speech object development .)

Edinburgh Speech Tools for speech synthesis

互联网著名国产软件小沁语音信使源代码。使用TAPI接口及语音Modem，实现语音电话自动拨打，语音录制及播放，广泛适用于企事业广告营销。(The use of TAPI interface and voice Modem, realize voice telephone automatic dial, voice recording and playback, widely applicable to the business of advertising and marketing. )

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)

gmm模型下的语音情感识别系统，GMM只是一个数学模型，只是对数据形态的拟和，但是和你所看到的数据分布存在出入也是正常的，因为用EM估计GMM的那些参数时，一般假设我们所得到的数据是不完备的（也就是说假设我们看到的数据分布不是真正的分布，它在运算时把那部分丢失或者叫隐藏的数据“补”上了）(gmm model speech emotion recognition system, GMM is a mathematical model, but fitting the data form, but you can see there is access to the data distribution is normal, because those with the EM estimated GMM parameters, the general assumption we get the data is incomplete (that is, assuming we see the distribution of the data distribution is not real, it is part of the operation at the time that is lost or hidden data is called " fill" on the))