Subband DCT and EMD Based Hybrid Soft Thresholding for Speech Enhancement

This paper presents a two-stage soft thresholding algorithm based on discrete cosine transform (DCT) and empirical mode decomposition (EMD). In the first stage, noisy speech is decomposed into eight frequency bands and a specific noise variance is calculated for each one. Based on this variance, each band is denoised using soft thresholding in DCT domain. The remaining noise is eliminated in the second stage through a time domain soft thresholding strategy adapted to the intrinsic mode functions (IMFs) derived by applying EMD on the signal obtained from the first stage processing. Significantly better SNR improvement and perceptual speech quality results for different noise types prove the superiority of the proposed algorithm over recently reported techniques.

Download Full-text

Preterm-Term Birth Classification Using EMD-Based Time-Domain Features of Single-Channel Electrohysterogram Data

10.21203/rs.3.rs-570938/v1 ◽

2021 ◽

Author(s):

Suparerk Janjarasjitt

Keyword(s):

Preterm Birth ◽

Time Domain ◽

Single Channel ◽

Support Vector ◽

Computational Results ◽

Intrinsic Mode Functions ◽

Term Birth ◽

Emg Signal ◽

Mode Decomposition ◽

The Time Domain

Abstract The preterm birth anticipation is a crucial task that can reduce the rate of preterm birth and also the complications of preterm birth. Electrohysterogram (EHG) or uterine electromyogram (EMG) data have been evidenced that they can provide an information useful for preterm birth anticipation. Four distinct time-domain features, i.e., mean absolute value, average amplitude change, difference absolute standard deviation value, and log detector, commonly applied to EMG signal processing are applied and investigated in this study. A single-channel of EHG data is decomposed into its constituent components, i.e., intrinsic mode functions, using empirical mode decomposition (EMD) before their time-domain features are extracted. The time-domain features of intrinsic mode functions of EHG data associated with preterm and term births are applied for preterm-term birth classification using support vector machine (SVM) with a radial basis function. The preterm-term classifications are validated using 10-fold cross validation. From the computational results, it is shown that the excellent preterm-term birth classification can be achieved using a single-channel of EHG data. The computational results further suggest that the best overall performance on preterm-term birth classification is obtained when thirteen (out of sixteen) EMD-based time-domain features are applied. The best accuracy, sensitivity, specificity, and F1-score achieved are, respectively, 0.9382, 0.9130, 0.9634, and 0.9366.

Download Full-text

Intrinsic Pattern Preserving Boundary Treatment Method for Empirical Mode Decomposition

Advances in Data Science and Adaptive Analysis ◽

10.1142/s2424922x1650008x ◽

2016 ◽

Vol 08 (02) ◽

pp. 1650008

Author(s):

Donghoh Kim ◽

Hee-Seok Oh

Keyword(s):

Empirical Mode Decomposition ◽

Treatment Method ◽

Intrinsic Mode Functions ◽

Boundary Problems ◽

Second Stage ◽

Numerical Studies ◽

Mode Decomposition ◽

Boundary Treatment ◽

Predicted Values ◽

Two Stages

Empirical mode decomposition (EMD) is a procedure that decomposes a signal into so-called intrinsic mode functions (IMFs) according to the levels of local frequency. Due to its robustness to nonlinear and nonstationary signals, EMD has been widely used in various fields. However, EMD suffers from boundary problems severely. In this paper, an efficient method for boundary treatment is proposed. The proposed method consists of two stages. In the first stage, regression models are adapted to reproduce the intrinsic sinusoid pattern of a signal. Based on predicted values, the signal is extended beyond the boundaries in the second stage. Results from numerical studies including simulation study and a noisy signal analysis demonstrate that the proposed method alleviates the boundary problem and hence provides more accurate decomposition results.

Download Full-text

Applied-Information Technology with Speech Enhancement Based on EMD and MF

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.1046.384 ◽

2014 ◽

Vol 1046 ◽

pp. 384-387

Author(s):

Jin Li ◽

Kun Shen

Keyword(s):

Information Technology ◽

Speech Enhancement ◽

Time Domain ◽

Improved Method ◽

Noisy Environments ◽

Structuring Element ◽

Mode Function ◽

Mode Decomposition ◽

Threshold Processing

Aiming at traditional methods cannot get good performance in noisy environments, an improved method for speech enhancement based on Empirical Mode Decomposition (EMD) and Morphology Filtering (MF) was proposed. The method firstly uses EMD to obtain Intrinsic Mode Function (IMF) and for hard threshold processing, then selects appropriate structuring element to construct MF for filtering processing in remaining IMFs. Finally, speech enhancement signal is reconstructed for each IMFs. Experimental results show that the proposed method for speech enhancement has better de-noising effect by comparing time-domain waveform and spectrogram. Moreover, the quality of reconstructed speech enhancement signal has been significantly improved.

Download Full-text

Speech Enhancement Based on EMD and Wavelet Threshold in Noisy Environments

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.989-994.3654 ◽

2014 ◽

Vol 989-994 ◽

pp. 3654-3657

Author(s):

Zi Qin Chen ◽

De Xiang Zhang ◽

Da Ling Yuan

Keyword(s):

Speech Enhancement ◽

Speech Processing ◽

Intrinsic Mode Functions ◽

Noisy Environments ◽

Mode Decomposition ◽

Limited Function ◽

Band Limited ◽

A New Technique ◽

Band Limited Function ◽

Target Characteristics

Speech enhancement is crucial for speech recognition accuracy. How to eliminate the effect of the noise constitutes a challenging problem in speech processing. This paper presents a new technique for speech enhancement in a noisy environment based on the empirical mode decomposition (EMD) algorithm and wavelet threshold. With the EMD, the noise speech signals can be decomposed into a sum of the band-limited function called intrinsic mode functions (IMFs), which is a zero-mean AM-FM component. Then wavelet threshold of the IMF components can be used to eliminate the effect of the noise for speech enhancement. Experimental results show that the proposed speech enhancement by de-noising algorithm is possible to achieve an excellent balance between suppresses noise effectively and preserves as many target characteristics of original signal as possible.

Download Full-text

Fault Diagnosis of Hoist Gearbox Based on Time-Domain Analysis of EMD and Fuzzy Clustering

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.328-330.1717 ◽

2011 ◽

Vol 328-330 ◽

pp. 1717-1720

Author(s):

Zi Gui Li ◽

Bi Juan Yan

Keyword(s):

Fault Diagnosis ◽

Fuzzy Clustering ◽

Time Domain ◽

Time Domain Analysis ◽

Domain Analysis ◽

Intrinsic Mode Functions ◽

Normal Surface ◽

Surface Fatigue ◽

Mode Decomposition ◽

The Time Domain

In this paper, the method of combining the time-domain analysis of empirical mode decomposition (EMD) and fuzzy clustering is explored for the hoist gearbox fault diagnosis. Firstly, it adopts the EMD technique to decompose the signal of vibration. With it, any complicated dataset can be decomposed into a finite and often small number of intrinsic mode functions (IMFs). Then a number of IMFs containing main fault information were selected, from which time domain feature parameters-- variance and kurtosis coefficient were extracted. At last, fuzzy clustering is used to diagnose and identify the kind of fault. The numerical simulation and the analysis of the response signal data from the hoist gearbox show that the method is effective at discriminating the three condition of the gear, i.e. the normal, surface fatigue pitting and cracked tooth.

Download Full-text

Single Channel Speech Enhancement Using Adaptive Soft-Thresholding with Bivariate EMD

ISRN Signal Processing ◽

10.1155/2013/724378 ◽

2013 ◽

Vol 2013 ◽

pp. 1-8 ◽

Cited By ~ 5

Author(s):

Md. Ekramul Hamid ◽

Md. Khademul Islam Molla ◽

Xin Dang ◽

Takayoshi Nakai

Keyword(s):

Speech Enhancement ◽

Speech Signal ◽

Single Channel ◽

Complex Signal ◽

Intrinsic Mode Functions ◽

Noisy Speech ◽

Mode Decomposition ◽

Data Adaptive ◽

Improved Performance ◽

Short Time

This paper presents a novel data adaptive thresholding approach to single channel speech enhancement. The noisy speech signal and fractional Gaussian noise (fGn) are combined to produce the complex signal. The fGn is generated using the noise variance roughly estimated from the noisy speech signal. Bivariate empirical mode decomposition (bEMD) is employed to decompose the complex signal into a finite number of complex-valued intrinsic mode functions (IMFs). The real and imaginary parts of the IMFs represent the IMFs of observed speech and fGn, respectively. Each IMF is divided into short time frames for local processing. The variance of IMF of fGn calculated within a frame is used as the reference term to classify corresponding noisy speech frame into noise and signal dominant frames. Only the noise dominant frames are soft-thresholded to reduce the noise effects. Then, all the frames as well as IMFs of speech are combined, yielding the enhanced speech signal. The experimental results show the improved performance of the proposed algorithm compared to the recently reported methods.

Download Full-text

ECG Noise Cancellation Based on Grey Spectral Noise Estimation

Sensors ◽

10.3390/s19040798 ◽

2019 ◽

Vol 19 (4) ◽

pp. 798 ◽

Cited By ~ 7

Author(s):

Shing-Hong Liu ◽

Cheng-Hsiung Hsieh ◽

Wenxi Chen ◽

Tan-Hsu Tan

Keyword(s):

Empirical Mode Decomposition ◽

Noise Cancellation ◽

Noise Estimation ◽

Body Motion ◽

Ecg Signal ◽

Intrinsic Mode Functions ◽

Reconstruction Process ◽

Second Stage ◽

Mode Decomposition ◽

Electrocardiogram Ecg

In recent years, wearable devices have been popularly applied in the health care field. The electrocardiogram (ECG) is the most used signal. However, the ECG is measured under a body-motion condition, which is easily coupled with some noise, like as power line noise (PLn) and electromyogram (EMG). This paper presents a grey spectral noise cancellation (GSNC) scheme for electrocardiogram (ECG) signals where two-stage discrimination is employed with the empirical mode decomposition (EMD), the ensemble empirical mode decomposition (EEMD) and the grey spectral noise estimation (GSNE). In the first stage of the proposed GSNC scheme, the input ECG signal is decomposed by the EMD to obtain a set of intrinsic mode functions (IMFs). Then, the noise energies of IMFs are estimated by the GSNE. When an IMF is considered as noisy one, it is forwarded to the second stage for further check. In the second stage, the suspicious IMFs are reconstructed and decomposed by the EEMD. Then the IMFs are discriminated with a threshold. If the IMF is considered as noisy, it is discarded in the reconstruction process of the ECG signal. The proposed GSNC scheme is justified by forty-three ECG signal datasets from the MIT-BIH cardiac arrhythmia database where the PLn and EMG noise are under consideration. The results indicate that the proposed GSNC scheme outperforms the traditional EMD and EEMD based noise cancellation schemes in the given datasets.

Download Full-text

Power System Stability Study with Empirical Mode Decomposition

Advanced Materials Research ◽

10.4028/www.scientific.net/amr.732-733.905 ◽

2013 ◽

Vol 732-733 ◽

pp. 905-908 ◽

Cited By ~ 2

Author(s):

Chia Liang Lu ◽

Pei Hwa Huang

Keyword(s):

Power System ◽

Empirical Mode Decomposition ◽

Time Domain ◽

Power System Stability ◽

Direct Analysis ◽

Stability Study ◽

System Stability ◽

Intrinsic Mode Functions ◽

Time Frequency ◽

Mode Decomposition

Low frequency oscillations (LFO) reflect the damping and the stability of a power system and is essentially non-stationary. The LFO is a composite response of various oscillation modes and of which the frequency may be changing with time; thus, direct analysis of such time-domain responses is difficult. The main purpose of this paper is to apply the method of empirical mode decomposition (EMD) to the study of power system stability. First the method of EMD is to expand the time-domain responses under study into multiple intrinsic mode functions (IMFs). Then the 2D time-frequency information inherent in the response under study is obtained using the wavelet transform. The 2D time-frequency graph is further expanded into a 3D time-frequency-energy graph. Information from the 3D time-frequency graph is analyzed to determine those generators that have higher extent of oscillation involvement during the occurrence of LFO in the power system. The results from comparative analysis show that, at specific frequencies from LFOs, higher extent of oscillation involvement will reveal a greater factor of involvement in the frequency domain behavior.

Download Full-text