Description: Rainer martin的早期文章,用于语音增强的噪声估计算法,对于最小统计量的提出有参考意义-Rainer martin early articles, speech enhancement for noise estimation algorithm, for the smallest of the proposed statistics are useful Platform: |
Size: 327680 |
Author:张小雨 |
Hits:
Description: 实验结果表明该算法对非冲击噪声效果良好。
lx_main为主程序,NS_lxwz为噪声估计模块,lxG_wz为增益估计模块,noise_sound为测试用含噪语音。
算法原理参见本人的文章《一种引入延迟的语音增强算法》。-Experimental results show that the algorithm for non-impulsive noise effect is good. lx_main the main program, NS_lxwz for the noise estimation module, lxG_wz for the gain estimation module, noise_sound to test the noisy speech. Algorithm theory see my article " delay the introduction of a speech enhancement algorithm." Platform: |
Size: 159744 |
Author:刘翔 |
Hits:
Description: This file is about, Relaxed Statistical Model for Speech
Enhancement and A Priori SNR Estimation Platform: |
Size: 510976 |
Author:Anusha |
Hits:
Description: 音频处理算法:语音增强技术是语音信号处理技术中的一门关键技术,本例是一种较为经典的用来估计噪声的算法程序,其实现较为简单,便于初学者学习,改程序已经通过仿真实验,并且实验结果用在方便哦的论文中。-Speech Enhancement technology is voice signal processing technology in a key technology, in this case is a more classical algorithm used to estimate the noise, its implementation is simple, easy for beginners to learn, to change the program has been through simulation experiments, and experiments results of paper in a convenient oh Platform: |
Size: 1024 |
Author:王 |
Hits:
Description: 使用matlab对基于高斯模型下的获取幅度估计的语音增强算法进行的仿真,效果较维纳算法和谱减算法有一定的改善。-Use matlab Gaussian model based on the estimated rate of access to speech enhancement algorithm for the simulation, the effect is less than the Wiener algorithm and the spectral algorithm has some improvement. Platform: |
Size: 1024 |
Author:王 |
Hits:
Description: 用于语音增强中的谱估计基本方法,针对音调较低,声音较沉闷的语音可以选用改进型的谱估计方法。-Spectrum estimation for speech enhancement in the basic method for tone lower than the dull sound of the voice can choose to use the improved spectral estimation method. Platform: |
Size: 1024 |
Author:xiaochong |
Hits:
Description: 马丁噪声估计算法,多用于谱减法,基于统计模型等语音增强算法中。-martin noise estimation,Used for spectral subtraction, based on statistical models, such as speech enhancement algorithm. Platform: |
Size: 1024 |
Author:尹海明 |
Hits:
Description: Speech processing applications such as speech enhancement and speaker identification rely on the estimation of relevant parameters from the speech signal. These
parameters must often be estimated from noisy observations since speech signals are
rarely obtained in ‘clean’ acoustic environments in the real world. As a result, the
parameter estimation algorithms we employ must be robust to environmental factors
such as additive noise and reverberation. In this work we derive and evaluate approximate Bayesian algorithms for the following speech processing tasks: 1) speech
enhancement 2) speaker identification 3) speaker verification and 4) voice activity
detection. Platform: |
Size: 1728512 |
Author:an mchol |
Hits:
Description: The material in this work is organized into five chapters, including this one. The
focus is on the time-domain algorithms for both the single and multiple microphone
cases. The work discussed in these chapters is as follows.
In Chap. 2, we study the noise reduction problem with a single microphone by
using a filtering vector for the estimation of the desired signal sample.
Chapter 3 generalizes the ideas of Chap. 2 with a rectangular filtering matrix for
the estimation of the desired signal vector.
In Chap. 4, we study the speech enhancement problem with a microphone array
by using a long filtering vector.
Finally, Chap. 5 extends the results of Chap. 4 with a rectangular filtering matrix Platform: |
Size: 506880 |
Author:infinity |
Hits:
Description: 这是目前传统单通道语音增强中效果最好的算法,作者Iseal Cohen大神,采用基于最小均方误差MMSE准则,代码里,噪声估计由最初的MCRA更新为效果更好的IMCRA。(This is the most effective algorithm for traditional single channel speech enhancement. The author, Iseal Cohen great God, uses the minimum mean square error MMSE criterion. In the code, the noise estimation is updated from the original MCRA to the better IMCRA.) Platform: |
Size: 10240 |
Author:Ahu_Lay
|
Hits: