搜索资源列表
NLMSspeechenhancement.rar
- 双通道语音增强算法，消除环境噪声。采用归一化自适应方法，噪声抵消10dB，语音保持较好可懂度。,Dual-channel speech enhancement algorithm, to eliminate environmental noise. The use of normalized adaptive methods, noise cancellation 10dB, to maintain good voice intelligibility.
MMSE
- 最小均方误差估计（MMSE）语音增强算法，算法简单有效，无残余音乐噪声-Minimum mean square error estimation (MMSE) speech enhancement algorithm is simple and effective, non-residual musical noise
voice_activity_detection
- 传统的有声无声检测方法只能在相对安静的环境下（<20dB）进行。因此传统方法不能满足噪声环境下的检测要求；另一方面，一直以来，单通道语音增强没有取得突破性进展，一个很重要的原因是没有找到高效的有声无声检测方法。研究了一种有声/无声检测算法是根据信号低频带、高频带短时能量包络的动态变化范围来判别噪声和含噪语音。对每一帧输入信号，通过与预先设定的一系列门限值比较来判断该帧是信号帧还是噪声帧。本设计使用了这种检测方法，选用北京瑞泰创新科技有限公司开发CETECK-VC33-D板，配合自己制作的前
IEEEXplore[1]
- 一种改进的语音增强算法的 子空间上的先进。它可以减少残留噪声引起的 -An Improved Algorithm for Speech Enhancement Advanced sub-space. It can reduce the residual noise caused by
MMSE_LSA
- 基于短时对数谱的最小均方准则语音增强算法-Number spectrum of the minimum mean square criterion speech enhancement algorithm based on short
zhuanli_adaptive4ch
- 谱相减 ＋ 波束形成 + 后Wiene滤波 语音增强算法-Spectral subtraction+ beamforming+ the Wiene filtering speech enhancement algorithm
Boll
- 改进boll谱减法，语音增强效果有提高。与MMSE算法不相上下-Improved the boll spectral subtraction speech enhancement effect has been improved. Comparable with the MMSE algorithm
log_spec_super_gaussV1_1
- 对数谱MMSE算法与谱减法的比较 可以加深对各种语音增强算法的优劣性的理解，便于学习-MMSE algorithm to compare the number of spectral subtraction spectrum
Untitled3
- 描写了语音增强算法中的谱减法，从有噪声的语音信号中提取尽可能纯净的原始语音-Describe the spectral subtraction in speech enhancement algorithm, noisy speech signal as much as possible to extract the purity of the original speech
91655265ArrayToolbox
- GSC语音增强算法，基于matlab的仿真程序-GSC speech enhancement algorithm based on matlab simulation program
nlms
- nlms语音增强算法，基于matlab的仿真程序-NLMS, speech enhancement algorithm based on matlab simulation program
DDalgorithm
- DD算法实现语音增强，噪声谱估计，根据寂静帧和后面能量的对比来估计-DD algorithm speech enhancement, noise spectrum estimation, based on comparative silence frame and rear energy to estimate
improved-beamformer-algorithm
- 内容是基于一种改进的麦克风阵列固定波束形成的语音增强算法，通过MATLAB仿真，其信噪比与之前的算法相比，有较大提升-Content is based on an improved fixed beamforming microphone array speech enhancement algorithm, through MATLAB simulation, the signal-to-noise ratio compared with the previous algorithm, has
small-dictionary
- 基于小字典训练和过完备稀疏表示的语音增强算法-Small dictionary-based voice training and over-complete sparse representation of enhancement algorithm
文献
- 文献中描述了语音增强的概述，对目前的算法进行分类总结，是很好的理论学习资料(The literature describes the speech enhancement overview, the current algorithm for classification and summary, is a very good theoretical learning materials)
kalmanfilter
- 使用卡尔曼算法对输入的数据进行平滑滤波，减小噪声的影响，可适用于语音增强，运动轨迹平滑等(The Kalman algorithm is used to smooth the input data and reduce the influence of noise. It can be applied to speech enhancement, motion trajectory smoothing and so on.)
BeamF
- 麦克风阵列语音增强处理，采用固定波束形成算法实现(Microphone array speech enhancement processing is realized by fixed beamforming algorithm.)
yasuo
- 麦克风阵列增强入门资料，麦克风阵列语音增强算法研究(Introduction to Microphone Array Enhancement and Research on Microphone Array Speech Enhancement Algorithms)
实验
- 此代码为传统语音增强算法的对比，包括谱减法、维纳滤波，自适应滤波，适合语音信号初学者使用。(This code is a comparison of traditional speech enhancement algorithms, including spectral subtraction, Wiener filtering, adaptive filtering, suitable for beginners to use voice signals.)
se_relativisticgan-master
- 基于gan的语音增强算法; 还包括原始gan用于语音增强的网络，还有wgan等(Speech enhancement based on GAN; Also includes the original GAN for voice enhancement network, wGAN and so on)