搜索资源列表
noiseshuaijian.zip
- 语音增强算法，该算法描述了经典谱减法的计算流程，结果用语谱图加以显示。,Speech enhancement algorithm, which describes the classical spectral subtraction in the calculation process, the results of the term spectrum to be displayed.
HTK
- HTK WIN下编译好的 可以用于语音识别 与MATLAB可以调用-The HTK WIN Next compiled for speech recognition and MATLAB can call
lpcc-matlab
- matlab程序，从语音中提取lpcc，可以修改维数，带详细中文解释，方便阅读。-matlab program, extracted from voice lpcc, you can modify the dimension, with a detailed explanation of the Chinese, and easy to read.
yinpinjiami
- 基于henon序列音频加密实验，1．构造有限长度的语音信号序列y(n)；2．通过4Matsumoto-Chua-kobayashi模型产生超混沌序列vn；3．将超混沌序列掩盖信号序列并获得加密信号序列，然后通过信道传输出去；4．接受方受到信号后采用超混沌信号序列去掩盖获得原信号序列；5．将实现方案采用Matlab语言编程并仿真正确； -Henon sequence based audio encryption experiment 1. Construct a finite-length s
hmm
- 功能函数，用于matlab的仿真和实现，可用于语音识别和语音处理-Functions for matlab simulation and implementation, can be used for speech recognition and speech processing
master2
- 基于GMM-HMM的语音识别程序,自带训练和测试数据，使用的是MATLAB。-GMM-HMM-based speech recognition program, comes with the training and test data, using MATLAB.
randlsbget
- 语音信息隐藏的matlab实现，有详细步骤，可以参考，有意义-the implementation of Speech information hiding by matlab , there are detailed steps
ierycoding
- 语音信号子带编码源程序,用matlab做的 不错的源码 很好-Speech signal source code, subband coding matlab do good source is very good
hopuy.
- 语音信号处理前分离信号噪声,matlab环境下运行,使信号处理效果更好-Speech signal processing before the separation of signal to noise, running under the matlab environment, make the signal processing effect is better
oource
- 语音信号子带编码源程序,用matlab做的 不错的源码 很好-Speech signal source code, subband coding matlab do good source is very good
Write
- 自己编写的用谱相减,最小均方和维纳滤波实现语音增强的matlab文件(Write your own using spectrum subtraction, wiener filtering and minimum mean square of speech enhancement matlab files)
uery
- 语音信号子带编码源程序,用matlab做的 不错的源码 很好(Speech signal source code, subband coding matlab do good source is very good)
awkv958
- GSM全速率语音突发的的发送端总程序,其中包括预编码,卷积编码,交织,调制等步骤的matlab程序,内付详细说明文档()
dtw
- 可以用来实现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。(Can be used to implement DTW)