资源列表
melbankm
- matlab编写，求mel滤波器矩阵的系数-Matlab prepared for mel filter coefficient matrix
lbg
- 矢量量化lbg算法的c语言实现。对训练序列实时N维矢量量化。适用于语音信号处理。-VQ lbg algorithm c language. Real-time vector quantization of the training sequence.
SpeechCore
- This function accepts the feature vector matrix as input and ％returns a cesptral mean normalised version of the input matrix
A_MFCC_CMS_GM
- MFCC, cepstral mean substraction and Gaussian mixture for Speaker ID
FFT
- fourier fast trasform for speech analisys
lpc_vq
- linear prediction coding for speech coding and compression. code not my me.
enframe
- 它是基于matlab的语音分帧的代码，相信对大家理解有帮助。-It is based on the matlab voice sub-frames of the code, I believe we understand help.
matlab-km
- this k-means clustering algorithm codes in matlab, and it is for beginners-this is k-means clustering algorithm codes in matlab, and it is for beginners
Desktop4
- mfcc3 program for new way
VB_example
- 利用ms speech sdk 开发的语音识别应用程序之Hello world-Hello world of vb speech recognition programming
yuyinshibie
- 输入测试语音特征，应用已训练好的网络权值，可以进行正确的语音识别。-speech recognition
DTW
- 在孤立词语音识别中，最为简单有效的方法是采用DTW（Dynamic Time Warping，动态时间归整）算法，该算法基于动态规划（DP）的思想，解决了发音长短不一的模板匹配问题，是语音识别中出现较早、较为经典的一种算法，用于孤立词识别。HMM算法在训练阶段需要提供大量的语音数据，通过反复计算才能得到模型参数，而DTW算法的训练中几乎不需要额外的计算。所以在孤立词语音识别中，DTW算法仍然得到广泛的应用。-In isolated word speech recognition, the mos