speech-signal-processing

attention model for saliency detection.

MFCC：Mel频率倒谱系数的缩写。Mel频率是基于人耳听觉特性提出来的，它与Hz频率成非线性对应关系。Mel频率倒谱系数(MFCC)则是利用它们之间的这种关系，计算得到的Hz频谱特征。(Speech signal feature extraction)

最经典的基于ICA实现的语音信号的采集、随机混合，再通过盲分离将混合后的语音信号分离(The most classical ICA-based speech signal acquisition, random mixing, and then the mixed speech signal is separated by blind separation.)

LMS自适应滤波算法是很常见的滤波算法。本代码包括常用的LMS自适应滤波算法，如基本LMS算法，解相关LMS算法，滤波型LMS算法，变换域LMS算法等。(LMS adaptive filtering algorithm is a common filtering algorithm. This code, including commonly used LMS adaptive filtering algorithms, such as the basic LMS algorithm, decorrelation LMS algorithm, filter-based LMS algorithm, transform domain LMS algorithm.)

uds 诊断主机模式，采用C语言作为开发语音。( This is a platform agnostic C library that implements the Unified Diagnostics Services protocol for automotive electronics. UDS is documented in ISO 14229 and is the underpinning for the more well-known On-board Diagnostics (OBD) standard. The library currently supports UDS running over CAN (ISO 15765-4), which uses the ISO-TP (ISO 15765-2) protocol for message framing. This library doesn t assume anything about the source of your diagnostic message requests or underlying interface to the CAN bus)

西奥多里蒂斯著，李晶皎译 本书系统阐述了模式识别的原理与方法，并在此基础上介绍了模式识别的应用。全书分为：基础部分和应用部分：基础部分主要包括统计模式识别、模糊模式识别、神经网络模式识别等内容；应用部分有车牌识别和语音识别。 (This paper discussed the principles and methods of pattern recognition, and based on this, the application of pattern recognition. Encyclopedia is divided into: basic part and application part: basic parts mainly include statistical pattern recognition, the fuzzy pattern recognition, neural network pattern recognition content Application part have license plate identification and voice recognition.)

DSP codes,Radial Basis functions, sum of sinusoidal signals,Histogram

语音识别功能，可以直接说话进行识别，然后转换成文字(Speech recognition)

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.)

块和子带自适应滤波器matlab源文件，内有详细说明及运行结果图(Block and Subband Adaptive Filters)