speech-recognition

SVM分类器，针对语音信号的分类识别应用。内容非常多，大家可以筛选着利用。(SVMclassification)

声纹识别和语音识别，语音识别的模块使用的讯飞云。声纹识别的代码则是参考了Dake前辈的代码。在此提出感谢。(Voiceprint recognition and speech recognition, voice recognition module for use hearing Fei. Voiceprint identification code is a reference to the Dake predecessors code. Presented in this thanks.)

speech specgram code

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.)

MFCC特征提取实例MFCC特征提取实例MFCC特征提取实例MFCC特征提取实例(Examples MFCC feature extraction)

端点检测算法采用现阶段比较精确的时频方差检测法，模式识别和模板匹配采用的是改进后的DTW算法（主要是限定了DTW的搜索路径，进一步精确了DTW的平行四边形的形状，进一步的减少匹配中的参数存贮量以及多余的搜索路径）。(Endpoint detection algorithm uses more precise at this stage variance time-frequency detection, pattern recognition and template matching is based on the improved DTW algorithm (DTW is mainly limited to the search path, and further precision of the DTW parallelogram shape, and further reduction of matching the parameters in the storage volume as well as the extra search path).)

智能天线波束形成的LMS、SMI、RLS算法的程序(Smart antenna beamforming of LMS, SMI, RLS algorithm procedure)

本资料涵盖了几乎所有的语音增强方面的方法，主要有谱减法，听觉掩蔽，最小均方误差，维纳滤波以及一些非主流的方法，这些对于研究语音增强的人来说是很有帮助的(The data cover almost all aspects of speech enhancement methods, the main spectral subtraction, auditory masking, minimum mean square error, Wiener filtering as well as some non-mainstream approach, which for the study of speech enhancement is helpful for people who)

科大讯飞语音转文字源码是一个基于科大讯飞的语音转文字的例子源码，小编测试一下表示被着准确率惊呆了，之前小编用过其他的语音搜索类的工具，但是准确率总是不那么满意，用这个例子我试着读了杂志上的一段话，语速还略微快了一些，但是转换出来的文字只有两个字跟原文不一样，但是还是同音字。(IFLYTEK speech text source is a turn based on iFLYTEK speech text examples of the source code, test bit representation of the being accurate rate was shocked, before Xiaobian used other voice search tool, but the accuracy is always not so satisfactory, for this example I try to read a passage in the magazine, speed is slightly faster, but convert text only two words with the original text is not the same, but still homophones.)

语音信号的时域、频域分析,包括短时能量分析、短时平均过零率、自相关函数、短时平均幅度差函数等。(Time-domain speech signal, frequency domain analysis, including short-term energy analysis, the average short-term zero-crossing rate, autocorrelation function, such as short-time average magnitude difference function.)