scholarly journals Evaluating Performance of Compressive Sensing for Speech Signal with various Basis

2014 ◽  
Vol 94 (11) ◽  
pp. 23-26
Author(s):  
Desai Siddhi ◽  
Nakrani Naitik
Keyword(s):  
Compressive Sensing ◽  
Speech Signal

Compressive Sensing and Contourlet Transform Applications in Speech Signal

2020 ◽  
pp. 833-842
Author(s):  
Korla Ramya ◽  
Vijayasri Bolisetti ◽  
Durgesh Nandan ◽  
Sanjeev Kumar
Keyword(s):  
Compressive Sensing ◽  
Speech Signal ◽  
Contourlet Transform

Speech Enhancement based on Wiener Filter and Compressive Sensing

2016 ◽  
Vol 2 (2) ◽  
pp. 367
Author(s):  
Amart Sulong ◽  
Teddy Surya Gunawan ◽  
Mira Kartiwi
Keyword(s):  
Compressive Sensing ◽  
Speech Enhancement ◽  
Speech Signal ◽  
Signal To Noise Ratio ◽  
Algorithm Design ◽  
Wiener Filter ◽  
Noise Signal ◽  
High Signal ◽  
Noisy Speech ◽  
Performance Results

<p><em>In communication medium to satisfy the speech enhancement process by using differents methodologies and algoirthms are the key term in testing the system design well enough to produce the best performance results for the speech system. The Wiener filter is one of the classical algorithm that applied to speech process to avoid the noise attacking the speech signal. In other word, compressive sensing method by randomize measurement matrix are combined with the Wiener filter to analyse the noisy speech signal with less introduce to noise signal and producing high signal to noise ratio. The PESQ is used to measure the quality of the proposed algorithm design. As in the experimental results shows that, attacking of defferent noise environments in speech signal still effectively improve the performance of noisy speech with maintain the high score of the PESQ quality. </em><em></em></p>

scholarly journals Watermarking Based on Compressive Sensing for Digital Speech Detection and Recovery †

Sensors ◽  
2018 ◽  
Vol 18 (7) ◽  
pp. 2390 ◽  
Author(s):  
Wenhuan Lu ◽  
Zonglei Chen ◽  
Ling Li ◽  
Xiaochun Cao ◽  
Jianguo Wei ◽  
...  
Keyword(s):  
Deep Learning ◽  
Compressive Sensing ◽  
Speech Signal ◽  
Coincidence Problem ◽  
Speech Detection ◽  
Dct Domain ◽  
Enhancement Method ◽  
Dct Coefficients ◽  
Content Recovery

In this paper, a novel imperceptible, fragile and blind watermark scheme is proposed for speech tampering detection and self-recovery. The embedded watermark data for content recovery is calculated from the original discrete cosine transform (DCT) coefficients of host speech. The watermark information is shared in a frames-group instead of stored in one frame. The scheme trades off between the data waste problem and the tampering coincidence problem. When a part of a watermarked speech signal is tampered with, one can accurately localize the tampered area, the watermark data in the area without any modification still can be extracted. Then, a compressive sensing technique is employed to retrieve the coefficients by exploiting the sparseness in the DCT domain. The smaller the tampered the area, the better quality of the recovered signal is. Experimental results show that the watermarked signal is imperceptible, and the recovered signal is intelligible for high tampering rates of up to 47.6%. A deep learning-based enhancement method is also proposed and implemented to increase the SNR of recovered speech signal.

scholarly journals Compressed Sensing Speech Signal Enhancement Research

2017 ◽  
Vol 6 (1) ◽  
pp. 26
Author(s):  
Kuangfeng Ning ◽  
Guojun Qin
Keyword(s):  
Compressed Sensing ◽  
Compressive Sensing ◽  
Input Signal ◽  
Speech Signal ◽  
Signal To Noise Ratio ◽  
Sampling Theorem ◽  
Compressed Domain ◽  
Signal To Noise ◽  
Perceptual Evaluation

<span lang="EN-US">The proposed Compressive sensing method is a new alternative method</span><span lang="EN-US">, it is</span><span lang="EN-US"> used to eliminate noise from the input signal</span><span lang="EN-US">,</span><span lang="EN-US"> and the quality of the speech signal </span><span lang="EN-US">is </span><span lang="EN-US">enhance</span><span lang="EN-US">d</span><span lang="EN-US"> with fewer samples</span><span lang="EN-US">, thus it is</span><span lang="EN-US"> required for the reconstruction than needed in some of the methods like Nyquist sampling theorem. The basic idea is</span><span lang="EN-US"> that </span><span lang="EN-US">the speech signals are sparse in nature</span><span lang="EN-US">,</span><span lang="EN-US"> and most of the noise signals are non-sparse in nature, and Compressive </span><span lang="EN-US">S</span><span lang="EN-US">ensing</span><span lang="EN-US">(</span><span lang="EN-US">CS) eliminates the non-sparse components and </span><span lang="EN-US">it </span><span lang="EN-US">reconstructs only the sparse components of the input signal. Experimental results prove that the average segmental SNR (signal to noise ratio) and PESQ (perceptual evaluation of speech quality) scores are better in the compressed domain</span><span lang="EN-US">．</span>

Sign in / Sign up

Export Citation Format

Share Document