Neural FET small-signal modelling based on mel-frequency cepstral coefficients

Author(s):  
Rania R. Elsharkawy ◽  
Sayed El-Rabaie ◽  
Moataza Hindy ◽  
Reda S. Ghoname ◽  
Moawad I. Dessouky
2009 ◽  
Vol 18 ◽  
pp. 185-204 ◽  
Author(s):  
Rania Elsharkawy ◽  
S. El-Rabaie ◽  
M. Hindy ◽  
R. S. Ghoname ◽  
Moawad Ibrahim Dessouky

2010 ◽  
Vol 19 (08) ◽  
pp. 1835-1846 ◽  
Author(s):  
R. R. ELSHARKAWY ◽  
M. HINDY ◽  
S. EL-RABAIE ◽  
M. I. DESSOUKY

In this paper, a novel neural technique is proposed for FET small-signal modeling. This technique is based on the discrete cosine transform (DCT) and the Mel-frequency cepstral coefficients (MFCCs). The input data to traditional neural systems for FET small-signal modeling are the scattering parameters and the corresponding frequencies in a certain band, and the outputs are the circuit elements. In the proposed technique, the input data are considered random, and the MFCCs are calculated from these inputs and their DCT. The MFCCs are used to give a few features from the input random data sequence to be used for the training of the neural networks. The objective of using MFCCs is to characterize the random input sequence with features that are robust against measurement errors. The MFCCs extracted from the DCT of the inputs increase the robustness against measurement errors. There are two benefits that can be achieved using the proposed technique; a reduction in the number of neural inputs and hence a faster convergence of the neural training algorithm and a robustness against measurement errors in the testing phase. Experimental results show that the technique based on the DCT and MFCCs is less sensitive to measurement errors than using the actual measured scattering parameters.


2019 ◽  
Vol 34 (1) ◽  
pp. 361-371 ◽  
Author(s):  
Mandip Pokharel ◽  
Avishek Ghosh ◽  
Carl Ngai Man Ho

Author(s):  
Musab T. S. Al-Kaltakchi ◽  
Haithem Abd Al-Raheem Taha ◽  
Mohanad Abd Shehab ◽  
Mohamed A.M. Abdullah

<p><span lang="EN-GB">In this paper, different feature extraction and feature normalization methods are investigated for speaker recognition. With a view to give a good representation of acoustic speech signals, Power Normalized Cepstral Coefficients (PNCCs) and Mel Frequency Cepstral Coefficients (MFCCs) are employed for feature extraction. Then, to mitigate the effect of linear channel, Cepstral Mean-Variance Normalization (CMVN) and feature warping are utilized. The current paper investigates Text-independent speaker identification system by using 16 coefficients from both the MFCCs and PNCCs features. Eight different speakers are selected from the GRID-Audiovisual database with two females and six males. The speakers are modeled using the coupling between the Universal Background Model and Gaussian Mixture Models (GMM-UBM) in order to get a fast scoring technique and better performance. The system shows 100% in terms of speaker identification accuracy. The results illustrated that PNCCs features have better performance compared to the MFCCs features to identify females compared to male speakers. Furthermore, feature wrapping reported better performance compared to the CMVN method. </span></p>


Sign in / Sign up

Export Citation Format

Share Document