资源列表
MFCC.rar
- 噪声鲁棒性说话人识别语音高频加权MFCC提取各方面介绍,Noise robustness of speaker recognition, voice high-frequency weighted to introduce various aspects of MFCC extraction
hmm.rar
- hmm的实现，包括训练、解码、评估3个主要部分。压缩包中有hmm教程,hmm realization, including three main parts-training, decoding and evaluation . There is a Hmm course in the package.
martin2001.zip
- 经典噪声估计，用于speech enhancement,noise estimation algorithm in speech enhancement
dtw.rar
- 很有用的DTW音频识别算法，可执行，matlab实现。,A useful DTW audio recognition algorithm, writed by matlab. The code can be performed.
Inpainting
- implementation of the inpainting algorithm in the papers "Region Filling and Object Removal by Exemplar-Based Image Inpainting" and "Gradient based image completion by solving the Poisson equation"
time-frequency-Toolbox-
- MATLAB时频分析工具箱源代码，包括使用说明各种函数库-Toolbox MATLAB source code for time-frequency analysis
hmm
- 基于HMM的语音识别程序 很完整详细 里面包括10几个独立的小程序-HMM-based speech recognition procedures are complete details of which includes 10 small program that several independent
fenxing
- 为提高语音端点检测(VAD)在较低信噪比(10 dB)下的准确率,提出一种基于短时分形维数的改进算法。结合语音信号的特点,对2种常用的语音信号分形维数计算方法进行了比较和选择,同时采用动态跟随门限值实现语音端点的自适应检测。试验结果表明:对于信噪比6~10 dB的带噪语音,此方法可以实现整段语音的检测,而且具有一定的噪声鲁棒性,系统运行期间能够自适应调整门限值以适应环境噪声的变化,提高了VAD算法的准确率。这个是源码matlab。-In order to improve voice activi
rastaplp
- rastaplp m-file calculates RASTA-PLP coefficients, defaults to 1 if 0, just calculate PLP coefficients.-rastaplp m-file calculates RASTA-PLP coefficients, defaults to 1 if 0, just calculate PLP coefficients.
vad.rar
- 自己写的关于语音识别技术中预处理过程中端点检测技术源码，用的matlab语言,Wrote it myself on voice recognition technology in the pretreatment process endpoint detection technology source, using the matlab language
Sound.rar
- 一个C++的语音识别系统，可用，值得参考,Sound
VoiceSeperationML.rar
- 两路语音信号混合及解混的基于负熵的fastICA算法的MATLAB仿真实例,Two-way voice signals mixed and mixed solutions of negative entropy-based algorithm fastICA simulation of MATLAB