Speech Endpoint Detection Based on EMD and Spectral Entropy in Noisy Environments

2012 ◽  
Vol 229-231 ◽  
pp. 1296-1299 ◽  
Author(s):  
Yan Li Liu ◽  
De Xiang Zhang ◽  
Ming Wei Ji
Keyword(s):  
Speech Recognition ◽  
Recognition Accuracy ◽  
Endpoint Detection ◽  
Intrinsic Mode Functions ◽  
Noisy Environment ◽  
Spectral Entropy ◽  
Noisy Environments ◽  
Novel Approach ◽  
Mode Decomposition ◽  
Speech Endpoint Detection

Accurate endpoint detection is crucial for speech recognition accuracy. A novel approach that finds robust features for endpoint detection based on the empirical mode decomposition (EMD) algorithm and spectral entropy in a noisy environment is proposed. With the EMD, the noise signals can be decomposed into different numbers of sub-signals called intrinsic mode functions (IMFs), which is a zero-mean AM-FM component. Then spectral entropy can be used to extract the desired feature for IMF components. In order to show the effectiveness of the proposed method, we present examples showing that the new measure is more effective than traditional measures. The experiments show that the proposed algorithm can suppress different noise types with different SNR, and the algorithm is robust in the real signal tests.

Speech Stream Detection for Noisy Environments Based on Empirical Mode Decomposition

2013 ◽  
Vol 397-400 ◽  
pp. 2239-2242
Author(s):  
Qiang Tang ◽  
De Xiang Zhang ◽  
Qing Yan
Keyword(s):  
Empirical Mode Decomposition ◽  
Recognition Performance ◽  
Intrinsic Mode Functions ◽  
Noisy Environment ◽  
Spectral Entropy ◽  
Noisy Environments ◽  
New Approach ◽  
Speech Stream ◽  
Mode Decomposition ◽  
Teager Energy

A new approach for speech stream detection based on empirical mode decomposition (EMD) under a noisy environment is proposed. Accurate speech stream detection proves to significantly improve speech recognition performance under noise. The proposed algorithm relies on the Teager energy and spectral entropy characteristics of the signal to determine whether an input frame is speech or non-speech. Firstly, the noise signals can be decomposed into different numbers of sub-signals called intrinsic mode functions (IMFs) with the EMD. Then, spectral entropy is used to extract the desired feature for noisy IMF components and Teager energy is used to non-noisy IMF components. Finally, in order to show the effectiveness of the proposed method, we present examples showing that the new measure is more effective than traditional measures. The experiments show that the proposed algorithm can suppress different noise types with different SNR.

scholarly journals Speech Endpoint Detection in Noisy Environment Based on the Ensemble Empirical Mode Decomposition

2011 ◽  
Vol 2-3 ◽  
pp. 135-139
Author(s):  
Jing Jiao Li ◽  
Dong An ◽  
Jiao Wang ◽  
Chao Qun Rong
Keyword(s):  
Empirical Mode Decomposition ◽  
Speech Signal ◽  
Recognition System ◽  
Ensemble Empirical Mode Decomposition ◽  
Endpoint Detection ◽  
Noisy Environment ◽  
Low Snr ◽  
Voice Signal ◽  
Mode Decomposition ◽  
Speech Endpoint Detection

Speech endpoint detection is one of the key problems in the practical application of speech recognition system. In this paper, speech signal contained chirp is decomposed into several intrinsic mode function (IMF) with the method of ensemble empirical mode decomposition (EEMD). At the same time, it eliminates the modal mix superposition phenomenon which usually comes out in processing speech signal with the algorithm of empirical mode decomposition (EMD). After that, selects IMFs contained major noise through the adaptive algorithm. Finally, the IMFs and speech signal contained chirp are input into the independent component analysis (ICA) and pure voice signal is separated out. The accuracy of speech endpoint detection can be improved in this way. The result shows that the new speech endpoint detection method proposed above is effective, and has strong anti-noises ability, especially suitable for the speech endpoint detection in low SNR.

A Speech Endpoint Detection Based on Empirical Mode Decomposition and Average Magnitude Difference Function

2013 ◽  
Vol 325-326 ◽  
pp. 1649-1652
Author(s):  
Wei Wei Shi ◽  
Wei Hua Xiong ◽  
Yun Yun Chu ◽  
Yu Liu
Keyword(s):  
Empirical Mode Decomposition ◽  
Speech Signal ◽  
Endpoint Detection ◽  
Intrinsic Mode Functions ◽  
Difference Function ◽  
Mode Decomposition ◽  
The Impact ◽  
Speech Endpoint Detection ◽  
Magnitude Difference ◽  
Average Magnitude Difference Function

Speech endpoint detection plays an important role in speech signal processing. In this paper, a method of speech endpoint detection based on empirical mode decomposition is introduced for accurately detecting the speech endpoint. This method used in speech signal decomposition gets a set of intrinsic mode functions (IMF). An IMF which contained a lot of noise must be filtered, and the rest of IMFs can be reconstructed to a new speech signal. The speech endpoint is detected by average magnitude difference function precisely. Simulation experiments show that the method proposed in this paper can eliminate the impact of noise effectively and detect the speech signal endpoint accurately.

EMD fractal feature extraction technique in fingerprint of medicinal herbs

2015 ◽  
Vol 13 (04) ◽  
pp. 1550022
Author(s):  
Jianwei Du ◽  
Zhengguang Xu ◽  
Zhichun Mu ◽  
Yuan Yan Tang ◽  
Limin Cui ◽  
...  
Keyword(s):  
Recognition Rate ◽  
Low Frequency ◽  
Fractal Dimensions ◽  
Medicinal Herbs ◽  
The Novel ◽  
Intrinsic Mode Functions ◽  
Special Situation ◽  
Novel Approach ◽  
Mode Decomposition ◽  
Wavelet Features

This paper proposes the fractal features for glycyrrhiza fingerprint of medicinal herbs, to obtain the intrinsic mode functions (IMFs) from high to low frequency by using empirical mode decomposition (EMD). The EMD fractal features are extracted through computing the fractal dimensions of each IMF. The novel approach is applied to the recognition of the three types of glycyrrhiza fingerprints. Experiments show that EMD fractal features have better recognition rate than that of the traditional ones in the case of concentration-change, i.e. the number of peak and peak drift of sample which has slight changes. An existing method to extract the fractal features for fingerprint of medicinal herbs based on wavelet transform, which is called fractal-wavelet features, was presented. This method has anti-jamming property against the change of samples concentration. However, the recognition rate based on fractal-wavelet features is not satisfactory when fingerprint of medicinal herbs has some slight concentrations changes, the number of peak and peak drift of samples are processed in the special situation.

Quantifying the complexity of human colonic pressure signals using an entropy measure

2016 ◽  
Vol 61 (1) ◽  
pp. 127-132 ◽  
Author(s):  
Fei Xu ◽  
Guozheng Yan ◽  
Kai Zhao ◽  
Li Lu ◽  
Zhiwu Wang ◽  
...  
Keyword(s):  
Signal Analysis ◽  
Colonic Motility ◽  
Approximate Entropy ◽  
Entropy Measure ◽  
Intrinsic Mode Functions ◽  
Spectral Entropy ◽  
Mode Decomposition ◽  
Power Spectral ◽  
Reconstructed Signal ◽  
Non Stationary Signal

Abstract Studying the complexity of human colonic pressure signals is important in understanding this intricate, evolved, dynamic system. This article presents a method for quantifying the complexity of colonic pressure signals using an entropy measure. As a self-adaptive non-stationary signal analysis algorithm, empirical mode decomposition can decompose a complex pressure signal into a set of intrinsic mode functions (IMFs). Considering that IMF2, IMF3, and IMF4 represent crucial characteristics of colonic motility, a new signal was reconstructed with these three signals. Then, the time entropy (TE), power spectral entropy (PSE), and approximate entropy (AE) of the reconstructed signal were calculated. For subjects with constipation and healthy individuals, experimental results showed that the entropies of reconstructed signals between these two classes were distinguishable. Moreover, the TE, PSE, and AE can be extracted as features for further subject classification.

scholarly journals Efficient robust speech recognition with empirical mode decomposition using an FPGA chip with dual core

2020 ◽  
Vol 9 (2) ◽  
pp. 109
Author(s):  
Shing-Tai Pan ◽  
Ching-Fa Chen ◽  
Wen-Sin Tseng
Keyword(s):  
Speech Recognition ◽  
Embedded System ◽  
Empirical Mode Decomposition ◽  
Recognition System ◽  
Robust Speech Recognition ◽  
Intrinsic Mode Functions ◽  
Mode Decomposition ◽  
Original Speech ◽  
Dual Core ◽  
Computing Speed

The purpose of this paper is to accelate the computing speed of Empirical Mode Decomposition (EMD) based on multi-core embedded systems for robust speech recognition. A reconfigurable chip, Field Programmable Gate Array (FPGA), is used for the implementation of the designed system. This paper applies EMD to discompose some noised speech signals into several Intrinsic Mode Functions (IMFs). These IMFs will be combined to recover the original speech by multiplying their corresponding weights which were trained by Genetic Algorithms (GA). After applying Empirical Mode Decomposition (EMD), we obtain a cleaner speech for recognition. Due to the complexity of the computation of the EMD, a dual-core architecture of embedded system on FPGA is proposed to accelerate the computing speed of EMD for robust speech recognition. This will enhance the efficiency of embedded speech recognition system.

Sign in / Sign up

Export Citation Format

Share Document