scholarly journals Denoising Speech Based on Deep Learning and Wavelet Decomposition

2021 ◽  
Vol 2021 ◽  
pp. 1-10
Author(s):  
Li Wang ◽  
Weiguang Zheng ◽  
Xiaojun Ma ◽  
Shiming Lin
Keyword(s):  
Deep Learning ◽  
Frequency Band ◽  
Wavelet Decomposition ◽  
Amplitude Spectrum ◽  
Audio Signal ◽  
Output Amplitude ◽  
Target Network ◽  
Reduction Effect ◽  
The Time Domain ◽  
Short Time

The work proposed a denoising speech method using deep learning. The predictor and target network signals were the amplitude spectra of the wavelet-decomposition vectors of the noisy audio signal and clean audio signal, respectively. The output of the network was the amplitude spectrum of the denoised signal. Besides, the regression network used the input of the predictor to minimize the mean square error between its output and input targets. The denoised wavelet-decomposition vector was transformed back to the time domain by the output amplitude spectrum and the phase of the wavelet-decomposition vector. Then, the denoised speech was obtained by the inverse wavelet transform. This method overcame the problem that the frequency and time resolution of the short-time Fourier transform could not be adjusted. The noise reduction effect in each frequency band was improved due to the gradual reduction of the noise energy in the wavelet-decomposition process. The experimental results showed that the method has a good denoising effect in the whole frequency band.

scholarly journals Ultrasonic Assessment of Thickness and Bonding Quality of Coating Layer Based on Short-Time Fourier Transform and Convolutional Neural Networks

Coatings ◽  
2021 ◽  
Vol 11 (8) ◽  
pp. 909
Author(s):  
Azamatjon Kakhramon ugli Malikov ◽  
Younho Cho ◽  
Young H. Kim ◽  
Jeongnam Kim ◽  
Junpil Park ◽  
...  
Keyword(s):  
Fourier Transform ◽  
Coating Layer ◽  
Ultrasonic Pulse ◽  
Short Time Fourier Transform ◽  
Coating Materials ◽  
Time Frequency ◽  
High Attenuation ◽  
Bonding State ◽  
The Time Domain ◽  
Short Time

Ultrasonic non-destructive analysis is a promising and effective method for the inspection of protective coating materials. Offshore coating exhibits a high attenuation rate of ultrasonic energy due to the absorption and ultrasonic pulse echo testing becomes difficult due to the small amplitude of the second echo from the back wall of the coating layer. In order to address these problems, an advanced ultrasonic signal analysis has been proposed. An ultrasonic delay line was applied due to the high attenuation of the coating layer. A short-time Fourier transform (STFT) of the waveform was implemented to measure the thickness and state of bonding of coating materials. The thickness of the coating material was estimated by the projection of the STFT into the time-domain. The bonding and debonding of the coating layers were distinguished using the ratio of the STFT magnitude peaks of the two subsequent wave echoes. In addition, the advantage of the STFT-based approach is that it can accurately and quickly estimate the time of flight (TOF) of a signal even at low signal-to-noise ratios. Finally, a convolutional neural network (CNN) was applied to automatically determine the bonding state of the coatings. The time–frequency representation of the waveform was used as the input to the CNN. The experimental results demonstrated that the proposed method automatically determines the bonding state of the coatings with high accuracy. The present approach is more efficient compared to the method of estimating bonding state using attenuation.

scholarly journals fNIRS Signal Classification Based on Deep Learning in Rock-Paper-Scissors Imagery Task

2021 ◽  
Vol 11 (11) ◽  
pp. 4922
Author(s):  
Tengfei Ma ◽  
Wentian Chen ◽  
Xin Li ◽  
Yuting Xia ◽  
Xinhua Zhu ◽  
...  
Keyword(s):  
Deep Learning ◽  
Total Duration ◽  
Signal Classification ◽  
Imagery Task ◽  
Neural Activities ◽  
Computer Interfaces ◽  
Acquisition Device ◽  
The Time Domain ◽  
The Brain

To explore whether the brain contains pattern differences in the rock–paper–scissors (RPS) imagery task, this paper attempts to classify this task using fNIRS and deep learning. In this study, we designed an RPS task with a total duration of 25 min and 40 s, and recruited 22 volunteers for the experiment. We used the fNIRS acquisition device (FOIRE-3000) to record the cerebral neural activities of these participants in the RPS task. The time series classification (TSC) algorithm was introduced into the time-domain fNIRS signal classification. Experiments show that CNN-based TSC methods can achieve 97% accuracy in RPS classification. CNN-based TSC method is suitable for the classification of fNIRS signals in RPS motor imagery tasks, and may find new application directions for the development of brain–computer interfaces (BCI).

Optimized Mooring Line Simulation Using a Hybrid Method Time Domain Scheme

2014 ◽  
Author(s):  
Niels Hørbye Christiansen ◽  
Per Erlend Torbergsen Voie ◽  
Jan Høgsberg ◽  
Nils Sødahl
Keyword(s):  
Hybrid Method ◽  
Time Domain ◽  
Computation Time ◽  
Mooring Line ◽  
Mooring Lines ◽  
Fem Model ◽  
Input Variables ◽  
The Time Domain ◽  
The Cost ◽  
Short Time

Dynamic analyses of slender marine structures are computationally expensive. Recently it has been shown how a hybrid method which combines FEM models and artificial neural networks (ANN) can be used to reduce the computation time spend on the time domain simulations associated with fatigue analysis of mooring lines by two orders of magnitude. The present study shows how an ANN trained to perform nonlinear dynamic response simulation can be optimized using a method known as optimal brain damage (OBD) and thereby be used to rank the importance of all analysis input. Both the training and the optimization of the ANN are based on one short time domain simulation sequence generated by a FEM model of the structure. This means that it is possible to evaluate the importance of input parameters based on this single simulation only. The method is tested on a numerical model of mooring lines on a floating off-shore installation. It is shown that it is possible to estimate the cost of ignoring one or more input variables in an analysis.

Selection of random vibration theory procedures for the NGA-East project and ground-motion modeling

2021 ◽  
Vol 37 (1_suppl) ◽  
pp. 1420-1439
Author(s):  
Albert R Kottke ◽  
Norman A Abrahamson ◽  
David M Boore ◽  
Yousef Bozorgnia ◽  
Christine A Goulet ◽  
...  
Keyword(s):  
Ground Motion ◽  
Time Domain ◽  
Random Vibration ◽  
Amplitude Spectrum ◽  
Motion Modeling ◽  
Peak Response ◽  
Vibration Theory ◽  
Random Vibration Theory ◽  
The Time Domain ◽  
The Impact

Traditional ground-motion models (GMMs) are used to compute pseudo-spectral acceleration (PSA) from future earthquakes and are generally developed by regression of PSA using a physics-based functional form. PSA is a relatively simple metric that correlates well with the response of several engineering systems and is a metric commonly used in engineering evaluations; however, characteristics of the PSA calculation make application of scaling factors dependent on the frequency content of the input motion, complicating the development and adaptability of GMMs. By comparison, Fourier amplitude spectrum (FAS) represents ground-motion amplitudes that are completely independent from the amplitudes at other frequencies, making them an attractive alternative for GMM development. Random vibration theory (RVT) predicts the peak response of motion in the time domain based on the FAS and a duration, and thus can be used to relate FAS to PSA. Using RVT to compute the expected peak response in the time domain for given FAS therefore presents a significant advantage that is gaining traction in the GMM field. This article provides recommended RVT procedures relevant to GMM development, which were developed for the Next Generation Attenuation (NGA)-East project. In addition, an orientation-independent FAS metric—called the effective amplitude spectrum (EAS)—is developed for use in conjunction with RVT to preserve the mean power of the corresponding two horizontal components considered in traditional PSA-based modeling (i.e., RotD50). The EAS uses a standardized smoothing approach to provide a practical representation of the FAS for ground-motion modeling, while minimizing the impact on the four RVT properties ( zeroth moment, [Formula: see text]; bandwidth parameter, [Formula: see text]; frequency of zero crossings, [Formula: see text]; and frequency of extrema, [Formula: see text]). Although the recommendations were originally developed for NGA-East, they and the methodology they are based on can be adapted to become portable to other GMM and engineering problems requiring the computation of PSA from FAS.

Fault Diagnosis of Wet-Shift Clutch Based on STFT and Wavelet

2011 ◽  
Vol 301-303 ◽  
pp. 1560-1567 ◽  
Author(s):  
Man Chen ◽  
Biao Ma
Keyword(s):  
Fault Diagnosis ◽  
Fourier Analysis ◽  
Wavelet Decomposition ◽  
Pulse Signal ◽  
Background Signal ◽  
Vibration Signals ◽  
Diagnosis Method ◽  
Friction Plate ◽  
Short Time ◽  
Shift Clutch

The paper analyzes the failure mechanism of the wet shifting clutch, and puts forward the concept that the deformation of the clutch friction plate leads to the irregular collision between the driving and driven sides of disengaged clutch and accordingly forms the transient pulse signal; the short-time Fourier analysis on the vibration signals of failed clutch obtained via test proves such concept. The transient pulse signal in the relatively strong background signal is clearly extracted through the wavelet decomposition after zero setting, and an efficient wet shifting clutch fault diagnosis method is hereby formed.

scholarly journals A Multimodal Feature Fusion-Based Deep Learning Method for Online Fault Diagnosis of Rotating Machinery

Sensors ◽  
2018 ◽  
Vol 18 (10) ◽  
pp. 3521 ◽  
Author(s):  
Funa Zhou ◽  
Po Hu ◽  
Shuai Yang ◽  
Chenglin Wen
Keyword(s):  
Deep Learning ◽  
Fault Diagnosis ◽  
Real Time ◽  
Time Domain ◽  
Feature Fusion ◽  
Rotating Machinery ◽  
Time Domain Data ◽  
Diagnosis Method ◽  
The Time Domain ◽  
Potential Frequency

Rotating machinery usually suffers from a type of fault, where the fault feature extracted in the frequency domain is significant, while the fault feature extracted in the time domain is insignificant. For this type of fault, a deep learning-based fault diagnosis method developed in the frequency domain can reach high accuracy performance without real-time performance, whereas a deep learning-based fault diagnosis method developed in the time domain obtains real-time diagnosis with lower diagnosis accuracy. In this paper, a multimodal feature fusion-based deep learning method for accurate and real-time online diagnosis of rotating machinery is proposed. The proposed method can directly extract the potential frequency of abnormal features involved in the time domain data. Firstly, multimodal features corresponding to the original data, the slope data, and the curvature data are firstly extracted by three separate deep neural networks. Then, a multimodal feature fusion is developed to obtain a new fused feature that can characterize the potential frequency feature involved in the time domain data. Lastly, the fused new feature is used as the input of the Softmax classifier to achieve a real-time online diagnosis result from the frequency-type fault data. A simulation experiment and a case study of the bearing fault diagnosis confirm the high efficiency of the method proposed in this paper.

scholarly journals Barycenter Theorem in Phase Characteristics of Symmetric and Asymmetric Windows

Symmetry ◽  
2018 ◽  
Vol 10 (8) ◽  
pp. 329
Author(s):  
Jiufei Luo ◽  
Haitao Xu ◽  
Kai Zheng ◽  
Xinyi Li ◽  
Song Feng
Keyword(s):  
Frequency Estimation ◽  
Random Noise ◽  
Estimation Algorithm ◽  
Frequency Component ◽  
Phase Response ◽  
Numeric Simulation ◽  
Simulation Results ◽  
The Time Domain ◽  
Short Time ◽  
The Relationship

Asymmetric windows are of increasing interest to researchers because of the nonlinear and adjustable phase response, as well as alterable time delay. Short-time phase distortion can provide an essential improvement in speech coding, and also has better performance in speech recognition. The merits of asymmetric windows in the aspect of spectral behaviors have an important function in frequency component detection and parameter estimation. In this paper, the phase response of windows were further studied, and the phase characteristics of symmetric and asymmetric windows are described. The relationship between the barycenter of windows in the time domain, and the phase characteristic at the center of the main lobe in the frequency domain, was established. In light of the relationship, an improved version of the asymmetric window- based frequency estimation algorithm was proposed. The improved algorithm has advantages of straightforward implementation and computational efficiency. The numeric simulation results also indicate that the improved approach is more robust than the traditional method against additive random noise.

scholarly journals Time-Domain Circuit Modelling for Hybrid Supercapacitors

Energies ◽  
2021 ◽  
Vol 14 (20) ◽  
pp. 6837
Author(s):  
Fabio Corti ◽  
Michelangelo-Santo Gulino ◽  
Maurizio Laschi ◽  
Gabriele Maria Lozito ◽  
Luca Pugi ◽  
...  
Keyword(s):  
Time Domain ◽  
Black Box ◽  
Circuit Modeling ◽  
Hybrid Supercapacitor ◽  
Hybrid Supercapacitors ◽  
Linear Behavior ◽  
Modeling Techniques ◽  
The Time Domain ◽  
Short Time ◽  
Circuit Modelling

Classic circuit modeling for supercapacitors is limited in representing the strongly non-linear behavior of the hybrid supercapacitor technology. In this work, two novel modeling techniques suitable to represent the time-domain electrical behavior of a hybrid supercapacitor are presented. The first technique enhances a well-affirmed circuit model by introducing specific non-linearities. The second technique models the device through a black-box approach with a neural network. Both the modeling techniques are validated experimentally using a workbench to acquire data from a real hybrid supercapacitor. The proposed models, suitable for different supercapacitor technologies, achieve higher accuracy and generalization capabilities compared to those already presented in the literature. Both modeling techniques allow for an accurate representation of both short-time domain and steady-state simulations, providing a valuable asset in electrical designs featuring supercapacitors.

Sign in / Sign up

Export Citation Format

Share Document