338 Hits in 5.7 sec

An Algorithm Research on Speech Endpoints Detection and Its Implementation on DSP

2013 International Journal of Digital Content Technology and its Applications  
In order to improve the accuracy and real-time performance of speech endpoints detection, this paper puts forward speech endpoints detection algorithm based on wavelet analysis and fuzzy neural networks  ...  The algorithm firstly extracts the speech signal characteristic quantity based on wavelet analysis.  ...  In the development of speech endpoints detection, there are many classical algorithms. Speech endpoint detection based on short-time energy is a more classic method of feature extraction.  ... 
doi:10.4156/jdcta.vol7.issue8.3 fatcat:nkzxbvam45c5rjzbdyinm7bf4q

Shift endpoint trace selection algorithm and wavelet analysis to detect the endpoint using optical emission spectroscopy

Sihem Ben Zakour, Hassen Taleb
2016 Photonic Sensors  
Endpoint detection (EPD) is very important undertaking on the side of getting a good understanding and figuring out if a plasma etching process is done on the right way.  ...  Then, a sensitivity analysis of linear methods named principal component analysis (PCA) and factor analysis (FA), and the nonlinear method called wavelet analysis (WA) for both approximation and details  ..., which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a  ... 
doi:10.1007/s13320-016-0280-5 fatcat:cnng222qcranjfs2z6zg27fdjy

Speech detection in non-stationary noise based on the 1/f process

Fan Wang, Fang Zheng, Wenhu Wu
2002 Journal of Computer Science and Technology  
In this paper, an effective and robust active speech detection method is proposed based on the 1/f process technique for signals under non-stationary noisy environments.  ...  In our experiments, a 10-minute long speech with different types of noises ranging from 20dB to 5dB is tested using this new detection method.  ...  Future work will be done to combine the speech endpoint detection and the noise cancellation methods together based on 1/f process theory. s 's are a collection of mutually uncorrelated random variables  ... 
doi:10.1007/bf02949828 fatcat:2rafcg34kvbfxajhc4xaov2lry

Voice Activity Detection Algorithm based on Improved Radial Basis Function Neural Network

Bao-yuan Chen, Ya-qiong Lan, Jing-yang Liu, Zi-he Li, Xiao-yang Yu
2014 International Journal of Signal Processing, Image Processing and Pattern Recognition  
Voice activity detection (VAD) is the key of voice recognition, voice synthesis and speech-sound enhancement.For the sake of improve the accuracy and robustness of speech endpoint detection system.  ...  This paper presents a comprehensive detection method--Adaptive genetic algorithm radial basis function network.  ...  The speech endpoint detection methods can be divided into two categories: based on the characteristics of the signals and based on pattern recognition [2] .  ... 
doi:10.14257/ijsip.2014.7.5.16 fatcat:cnvtrqgccbhbdh7mejoyjy62rm

A Robust Multilevel Speech Verification With Wavelet Decomposition for Inadequate Training Data Sets of Mobile Device Systems

Kuo-Kun Tseng, Yang Zhang, K. L. Yung, W. H. Ip, Zhye Ou, Qi Na
2019 IEEE Access  
This paper proposes a novel architecture for speech verification using a multilevel method, which extracts feature parameters through a multiple wavelet transform for mobile phone voice.  ...  Most of the current speech verifications are based on long training data sets in order to achieve good results, and there are no good verification schemes in case of inadequate training data sets.  ...  For example, although the detection method based on the wavelet transform has a high detection accuracy, the speed of detection is very slow.  ... 
doi:10.1109/access.2018.2886428 fatcat:rhuxznvombeu3ndoiuc76mkm54

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

Jing Jiao Li, Dong An, Jiao Wang, Chao Qun Rong
2011 Advanced Engineering Forum  
Speech endpoint detection is one of the key problems in the practical application of speech recognition system.  ...  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.  ...  In order to check the effect of endpoint detection, the speech endpoint detection method based on the spectral entropy method was defined as compared method.  ... 
doi:10.4028/ fatcat:un56ole2jbabljrzlvq2l4au5a

HMM-Based Techniques for Speech Segments Extraction

Waleed H. Abdulla
2002 Scientific Programming  
This paper discusses two novel HMM based techniques that segregate a speech segment from its concurrent background.  ...  The first method can be reliably used in clean environments while the second method, which makes use of the wavelets denoising technique, is effective in noisy environments.  ...  One classical technique uses the energy level for the detection of endpoints of the speech segment.  ... 
doi:10.1155/2002/819429 fatcat:zobiglfl6rht7iczphl2qmjjxa

Robust endpoint detection algorithm based on the adaptive band-partitioning spectral entropy in adverse environments

Bing-Fei Wu, Kun-Ching Wang
2005 IEEE Transactions on Speech and Audio Processing  
Robust endpoint detection is one of the most important areas of speech processing. Generally, the feature parameters used for endpoint detection are highly sensitive to the environment.  ...  As a result, this study seeks the banded structure on speech spectrogram to distinguish a speech from a nonspeech, especially in adverse environments.  ...  His research interests include adaptive signal processing, speech enhancement, speech recognition, and wavelet analysis and applications.  ... 
doi:10.1109/tsa.2005.851909 fatcat:xceexethb5agnoyp26fohxcpie

A Multiscale Chaotic Feature Extraction Method for Speaker Recognition

Jiang Lin, Yi Yumei, Zhang Maosheng, Chen Defeng, Wang Chao, Wang Tonghan, Karthikeyan Rajagopal
2020 Complexity  
Then, we extracted the speech chaotic characteristics based on the nonlinear dynamic model, which helps to improve the discrimination of features.  ...  Under clean speech and noise speech conditions, the ERR value of our method is reduced by 13.94% and 26.5% compared with the state-of-the-art method, respectively.  ...  In this paper, endpoint detection adopts a double-threshold method based on zerocrossing rate and energy. e pre-emphasis is carried out by a first-order FIR high pass filter, and the pre-emphasis coefficient  ... 
doi:10.1155/2020/8810901 fatcat:6vst6abjknfz7eojwd36qw3jta

Data compression based on the cubic B-spline wavelet with uniform two-scale relation

S.K. Yang, C.H. Cooke
1996 Mathematical and computer modelling  
Data compression based on the cubic B-spline wavelet with uniform two-scale relation.  ...  Abstract-The aim of this paper is to investigate the potential artificial compression which can be achieved using an interval multiresolution analysis based on a semiorthogonal cubic B-spline wavelet.  ...  For cubic splines having two continuous derivatives, one should be able to detect discontinuities in a function and its first or second derivatives by performing a wavelet decomposition.  ... 
doi:10.1016/0895-7177(96)00030-1 fatcat:vcgc5ev265bxzdubq5upk2bmky

Adaptive wavelet thresholding with robust hybrid features for text-independent speaker identification system

Hesham A. Alabbasi, Ali M. Jalil, Fadhil S. Hasan
2020 International Journal of Electrical and Computer Engineering (IJECE)  
To eliminate the problem of additive noise and to achieve a high accuracy in speaker identification system a proposed algorithm for feature extraction based on speech enhancement and a combined features  ...  In this paper, a wavelet thresholding pre-processing stage, and feature warping (FW) techniques are used with two combined features named power normalized cepstral coefficients (PNCC) and gammatone frequency  ...  Improved spectral subtraction [6] based on two steps, speech activity detection (SAD) and noise amplitude spectral estimation; it was adopted based on frequency band variance to detect speech endpoints  ... 
doi:10.11591/ijece.v10i5.pp5208-5216 fatcat:6ct4fw67snfflkp3bcvelbgwzu

On the Use of Adaptive Fuzzy Wavelet Filter in the Speech Enhancement

Chih-Chia Yao, Ming-Hsun Tsai, Yuan-Tain Chang
2014 Journal of Computers  
This paper proposes an adaptive fuzzy wavelet filter that is based on a fuzzy inference system for enhancing speech signals and improving the accuracy of speech recognition.  ...  A hybrid filter is proposed here, which combines an adaptive fuzzy wavelet filter and the spectral subtraction method to filter contaminated signals.  ...  ADAPTIVE FUZZY WAVELET FILTER This section proposes an adaptive fuzzy wavelet filter based on a fuzzy inference system.  ... 
doi:10.4304/jcp.9.11.2501-2513 fatcat:aqltk4itxvbbnkzylqbe6jixr4

Adaptive Fuzzy Filter for Speech Enhancement [chapter]

Chih-Chia Yao, Ming-Hsun Tsai
2010 Lecture Notes in Computer Science  
In this paper an adaptive fuzzy filter, based on fuzzy system, is proposed for speech signal enhancement and automatic speech recognition accuracy.  ...  First an amplified voice activity detector is designed to improve performance on SNR lower than 5dB. Then an adaptive threshold decision module based on fuzzy inference system is proposed.  ...  In this paper we propose an adaptive fuzzy filter, based on wavelet thresholding, to improve performance in speech signal enhancement and automatic speech recognition accuracy.  ... 
doi:10.1007/978-3-642-12179-1_42 fatcat:s3sqmtr22fganc725eeuu35pvq

The separability analysis of pitch parameters in speaker recognition

Rong Rong He, Lan Tian, Fa Jiang Ma, J.C.M. Kao, W.-P. Sung
2016 MATEC Web of Conferences  
which the separability of the speaker is obvious.We do a speaker identification experiment with a small corpus,and the experiment results show that speaker identification based on the prosodic parameters  ...  Aseparability research of prosodic parameters in speaker recognition is studied.The space distribution of the prosodic parameters is drawn using the mean and variance of the speaker pitch parameters,through  ...  To accurately extract the signal, it is necessary to adopt appropriate detection algorithm. Currently, there is a variety of pitch detection method.  ... 
doi:10.1051/matecconf/20166304047 fatcat:ar36uzblubcn5h6bw62qaaff5q

NNGD algorithm for neural adaptive filters

D.P. Mandic
2000 Electronics Letters  
methods.  ...  However, we believe that, although we have presented just a heuristic approach to such a design, its strength lies in the fact that it opens new research possibilities towards finding more effective design  ...  Description of algorithm: The architecture of the endpoint detector proposed is based on a pattern recognition approach.  ... 
doi:10.1049/el:20000631 fatcat:cc7plltjmreqdgqio7hhf6dtcy
« Previous Showing results 1 — 15 out of 338 results