▍1. EM_init_kmeans
高斯混合模型参数初始化程序,在对高斯混合模型的建立之前采用KMEANS算法进行初始化(Gaussian mixture model parameter initialization procedure, in the Gaussian mixture model is initialized before the algorithm used KMEANS)
高斯混合模型参数初始化程序,在对高斯混合模型的建立之前采用KMEANS算法进行初始化(Gaussian mixture model parameter initialization procedure, in the Gaussian mixture model is initialized before the algorithm used KMEANS)
hmm文件时运用HMM算法实现噪声环境下语音识别的。其中vad.m是端点检测程序;mfcc.m是计算MFCC参数的程序;pdf.m函数是计算给定观察向量对该高斯概率密度函数的输出概率;mixture.m是计算观察向量对于某个HMM状态的输出概率,也就是观察向量对该状态的若干高斯混合元的输出概率的线性组合;getparam.m函数是计算前向概率、后向概率、标定系数等参数;viterbi.m是实现Viterbi算法;baum.m是实现Baum-Welch算法;inithmm.m是初始化参数;train.m是训练程序;main.m是训练程序的脚本文件;recog.m是识别程序。(hmm HMM algorithm file using speech recognition in noisy environments. Which is the endpoint detection process vad.m mfcc.m procedure is to calculate the MFCC parameters pdf.m function is calculated for a given observation vector of the Gaussian probability density function of output probability mixture.m is to calculate the observation vector for a HMM state output probability of observation vector is the number of Gaussian mixture per state output probability of the linear combination getparam.m before the calculation of the probability function, backward probability, calibration coefficients and other parameters viterbi.m is Viterbi algorithm implementation baum.m Baum-Welch algorithm to achieve inithmm.m is the initialization parameters train.m is the training program main.m training program is a script file recog.m is to identify procedures.)
蓝牙全双工语音和数据传输 蓝牙调制和跳频 蓝牙语音传输的相关程序(Bluetooth full duplex voice and data transfer Bluetooth modulation and frequency hopping Bluetooth voice transmission procedures)
语音处理的实用代码,完成耳语音的声韵分割功能(Practical speech processing code, complete the ear speech sound segmentation function)
不错的盲源分离工具箱,实现对声源的分离,包含多个子程序(a nice toolbox of blind source separation used to separate multiple sources)
speech recognition in matlab which will validate the speech
attention model for saliency detection.
观测实时模拟信号(语音)的频谱 用音频设备采集一段语音,将语音存为.wav格式。对wav文件作分段傅里叶变换分析。语音是分音节的,应把它分段分析,而且实际运用中的数字信号处理的FFT的点数是有限的,一般只能达到千点。用傅里叶反变换IFFT,从频域恢复信号。画出频谱图和语音波形图。 (Observing real-time analog signal (voice) of the spectrum collected with the audio devices section of voice, voice saved as. Wav format. Wav file of the sub-Fourier transform analysis to make. Speech is divided into syllables, it should be sub-analysis, and practical application of digital signal processing of the FFT points is limited, generally only reach 1000 points. Fourier inverse transform IFFT, to restore the signal from the frequency domain. Draw the spectrum map and voice waveform.)
基于LPC的语音合成,可以实现变调变速功能,附带注释便于学习。(LPC-based speech synthesis, can achieve modulation speed function, annotated easy to learn.)
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.)
这里面的代码是2DPCA的。这篇文章发表在PAMI上,Two-Dimensional PCA: A New Approach to Appearance-Based Face Representation and Recognition.google搜索已经引用681次。这里的代码是原作者的。这个分类不太好,语音识别是最近的,这是模式识别里的经典算法。(matlab code of 2DPCA)
一个可执行的HMM语音识别程序例程,实现了对10个数字音的识别程序,包含了HMM语音识别中的分段,MFCC特征提取,Baum-Welch训练,及Viterbi等算法,通过此例程可以很好的理解HMM的算法原理(An executable HMM-based 10 digits speech recogntion program example. this code zip file includes segmentation, MFCC feature extraction, Baum-Welch based re-estimation and Viterbi algorithm involved in HMM. it helps much better understand the HMM algorithm and its application for speech recogntion. )
采样小波包分解语音信号,分解为3层,并求出分解系数(Speech signal using wavelet packet decomposition,decomposing 6 level,and attain decomposition coefficient)
本代码是基于matlab实现的melp算法,编码速率为2.4kb/s(This code is based on the matlab implementation melp algorithms, coding rate of 2.4kb/s)