搜索资源列表
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
VoiceBox-matlab-tool
- 语音信号处理的matlab tool, 适用于语音信号处理相关算法研究参考，适合入门级别。 -Speech signal processing matlab tool, for voice signal processing algorithm reference for entry level.
adpcm
- 该算法是根据自适应差分脉冲编码调制原理编写的算法，该算法利用了语音信号样点间的相关性，并针对语音信号的非平稳特点，使用了自适应预测-The algorithm is based on principles of adaptive differential pulse code modulation written algorithm, which samples the voice signal using the correlation between, and for non-station
speech_denoising_lms
- 利用LMS算法来对语音信号进行去噪处理，同时含有关于LMS的文档资料-LMS algorithm to the speech signal to deal with the noise, and also has the documents about LMS
psola.rar
- 语音信号处理调节韵律的matlab程序，基音同步叠加算法PSOLA，,Speech Signal Processing, Pitch Synchronous Overlap Add algorithm PSOLA
关于AMR-WB标准的音频压缩编码的算法和思想
- 详细介绍了关于AMR-WB标准的音频压缩编码的算法和思想,Detailed introduction on the AMR-WB standard audio compression coding algorithms and thought
mllr
- 语音识别最大似然线性回归（MLLR）算法-Speech recognition maximum likelihood linear regression (MLLR) algorithm
humanearbasedonthespectral
- 改进的基于人耳掩蔽效应谱减语音增强算法,是谱减法的改进算法。-To improve the human ear based on the spectral masking by speech enhancement algorithm is to improve the algorithm of spectral subtraction.
mydtw
- 基于MFCC算法建立的参数模型，然后用DTW算法进行模型匹配，从而实现语音识别的作用~-MFCC algorithm based on parameters established model, and model matching using DTW algorithm to achieve the role of speech recognition ~
spectrumsub
- 实现在单信道语音信号处理中的降噪，用Spectral Subtraction算法实现的，内涵音频文件，代码中包括给纯净音频文件加噪声的部分。-Noise reduction in single-channel speech signal processing, using the Spectral Subtraction algorithm, the connotation of an audio file, the code in the package Including pure au
melp_fxp.tar
- 音频编解码 2400 bps MELP语音编解码器 定点算法-audio codec 2400 bps MELP speech codec algorithm sentinel
pujianfa
- 基于谱减法的语音增强算法，谱相减法及其变形方法总体上运算量较小，容易实时实现，增强效果也较好，是目前最常用的一种方法-Spectral subtraction based speech enhancement algorithm, spectral subtraction method and the deformation of the whole volume of the smaller operations, easy real-time implementation, and enha
2400 bps MELP语音编解码器_fixpoint.tar
- 2400 bps MELP语音编解码器 定点运算算法-2400 bps MELP speech codec algorithm sentinel Operational
BURG
- 本文件为语音识别lpc源码中的burg算法部分。该算法简单实用，适合于进行线性预测参数的计算-This document is for lpc-source speech recognition algorithm in the part of the burg. The algorithm is simple and practical, suitable for the calculation of the linear prediction parameters
dtw
- 语音识别，基于DTW算法的语音识别原理与实现，包含源程序，MATLAB-Speech recognition, speech recognition algorithm based on DTW Principle and Implementation, including source code, MATLAB
blindsourceseparation
- 固定步长和变步长批处理自然梯度语音盲分离算法，包括主程序和调用程序。-Fixed step and variable step-size batch natural gradient algorithm for blind separation of speech, including the main program and the calling program.
melp2
- 2400 bps MELP语音编解码定点算法-2400 bps MELP speech codec fixed-point algorithm
refsoft
- 语音压缩算法 ，C语言编写 多平台 -c code for speech code algorithm,and support multi-platform
txt2wav
- 一个模式识别中聚类算法的实现，可用于语音识别的-Clustering of a pattern recognition algorithm can be used for speech recognition
20080229123745538
- 一个模式识别中聚类算法的实现，可用于语音识别的-Clustering of a pattern recognition algorithm can be used for speech recognition