6,052 Hits in 5.4 sec

Experimentation Using Short-Term Spectral Features for Secure Mobile Internet Voting Authentication

Surendra Thakur, Emmanuel Adetiba, Oludayo O. Olugbara, Richard Millham
2015 Mathematical Problems in Engineering  
In this study, higher dimensions of each of the short-term features were reduced to an 81-element feature vector per Speaker using Histogram of Oriented Gradients (HOG) algorithm while neural network ensemble  ...  We propose a secure mobile Internet voting architecture based on the Sensus reference architecture and report the experiments carried out using short-term spectral features for realizing the voice biometric  ...  Conflict of Interests The authors declare that there is no conflict of interests regarding the publication of this paper. Acknowledgment  ... 
doi:10.1155/2015/564904 fatcat:jfoq6oun4rb4tk2kgcjedxupa4

Robust Speaker Identification Using Fusion of Features and Classifiers

Smarajit Bose, Amita Pal, Anish Mukherjee, Debasmita Das
2017 International Journal of Machine Learning and Computing  
Speaker identification using Gaussian Mixture Models (GMMs) based on Mel Frequency Cepstral Coefficients (MFCCs) as features, proposed by Reynolds (1995), is one of the most effective approaches available  ...  with principal component transformation and some robust estimation procedures, can be used to enhance significantly the performance of the MFCC-GMM speaker recognition systems, using the benchmark speech  ...  Hema Murthy of the Indian Institute of Technology Madras for her invaluable assistance in the form of useful discussions and for loan of software related to the MFCC-GMM speaker recognition system.  ... 
doi:10.18178/ijmlc.2017.7.5.635 fatcat:b5pifjnwbbhbhlia6nc45x2kei

Improved Language-Independent Speaker Identification in a Non-contemporaneous Setup

Smarajit Bose, Amita Pal, Anish Mukherjee, Debasmita Das
2020 International Journal of Machine Learning and Computing  
One of the most effective approaches available in the literature for Automatic Speaker Identification is based on Gaussian Mixture Models (GMMs) with Mel Frequency Cepstral Coefficients (MFCCs) as features  ...  The use of GMMs for modeling speaker identity is motivated by the interpretation that the Gaussian components represent some general speaker-dependent spectral shapes, and the capability of mixtures to  ...  Bose proposed the application of the PCT, ensemble methods and trimmed mean.  ... 
doi:10.18178/ijmlc.2020.10.5.984 fatcat:rowxvwp2lnavzgsu6qj7gzjmke

Ensemble based speaker recognition using unsupervised data selection

Chien-Lin Huang, Jia-Ching Wang, Bin Ma
2016 APSIPA Transactions on Signal and Information Processing  
This paper presents an ensemble-based speaker recognition using unsupervised data selection.  ...  First, without any auxiliary information, we use ensemble classifiers based on unsupervised data selection to make use of different acoustic characteristics of speech data.  ...  Using the LTF and the ensemble method decreases the amount of data for training, because the LTF provides the more compact feature and the ensemble method divides data into subsets.  ... 
doi:10.1017/atsip.2016.10 fatcat:jhfqksc42vce3kousfzi2gfu5a

Acoustic Model Ensembling Using Effective Data Augmentation for CHiME-5 Challenge

Feng Ma, Li Chai, Jun Du, Diyuan Liu, Zhongfu Ye, Chin-Hui Lee
2019 Interspeech 2019  
Different from the conventional data simulation methods, we use a signal processing method originally developed for channel identification to estimate the room impulse responses and then simulate the far-field  ...  In this study, we present five different kinds of robust acoustic models which take advantages from both effective data augmentation and ensemble methods to improve the recognition performance for the  ...  Different from these conventional simulation methods, we use a signal processing method originally developed for channel identification to estimate the room impulse responses and then simulate the far-field  ... 
doi:10.21437/interspeech.2019-2601 dblp:conf/interspeech/Ma0DLYL19 fatcat:kdmkv6lwljathjol6zwkx7xq5i

Robust Gender Identification using EMD-Based Cepstral Features

Ghasem Alipoor, Ehsan Samadi
2018 Asia-Pacific Journal of Information Technology and Multimedia  
In this paper, using the empirical mode decomposition (EMD), some new and improved mel-frequency cepstral coefficient (MFCC) features are developed to address the problem of robust speaker gender identification  ...  Automatic speaker gender identification is a field of research with numerous practical applications.  ...  Furthermore, as a pre-processing unit, gender identification can enhance the accuracy of some recognition models, e.g. within the speaker identification (Khelif, Mombrun et al. 2017) , speaker verification  ... 
doi:10.17576/apjitm-2018-0701-06 fatcat:ehh35ys7xbarzk4xdogfvbyh7i

Vowel-based Meeteilon dialect identification using a Random Forest classifier [article]

Thangjam Clarinda Devi, Kabita Thaoroijam
2021 arXiv   pre-print
Random forest classifier, a decision tree-based ensemble algorithm is used for classification of three major dialects of Meeteilon namely, Imphal, Kakching and Sekmai.  ...  Model has shown an average dialect identification performance in terms of accuracy of around 61.57%.  ...  and allowing us to use it for this analysis.  ... 
arXiv:2107.13419v1 fatcat:sg3ril6e3vhhner55va3bicsye

Perceptual Loss based Speech Denoising with an ensemble of Audio Pattern Recognition and Self-Supervised Models [article]

Saurabh Kataria, Jesús Villalba, Najim Dehak
2020 arXiv   pre-print
Our best model (PERL-AE) only uses acoustic event model (utilizing AudioSet) to outperform state-of-the-art methods on major perceptual metrics.  ...  Deep learning based speech denoising still suffers from the challenge of improving perceptual quality of enhanced signals.  ...  PERL leverages an ensemble of variety of opensource large-scale pre-trained speech models for deriving perceptual loss or Deep Feature Loss.  ... 
arXiv:2010.11860v1 fatcat:ofm2by6aynbixakhtck2cbpfni

Fusion methods for boosting performance of speaker identification systems

Gregory Ditzler, James Ethridge, Ravi P. Ramachandran, Robi Polikar
2010 2010 IEEE Asia Pacific Conference on Circuits and Systems  
Results using the King database show that both fusion methods lead to enhanced performance.  ...  Two important components of a speaker identification system are the feature extraction and the classification tasks.  ...  FUSION METHODS We present the use of ensemble based systems for data fusion [8] to augment the performance of the PFMRCEP, PFMRACW, and MRMFCC features.  ... 
doi:10.1109/apccas.2010.5774964 dblp:conf/apccas/DitzlerERP10 fatcat:qvgqkja53rbszd2dusfx6tyste

Ensemble of Jointly Trained Deep Neural Network-Based Acoustic Models for Reverberant Speech Recognition [article]

Jeehye Lee, Myungin Lee, Joon-Hyuk Chang
2016 arXiv   pre-print
Also, each model in the ensemble of DNN acoustic models is further jointly trained, including both feature mapping and acoustic modeling, where the feature mapping is designed for the dereverberation as  ...  In order to cope with a wide range of reverberations in real-world situations, we present novel approaches for acoustic modeling including an ensemble of deep neural networks (DNNs) and an ensemble of  ...  In a recent study, multiple datasets were generated through normalized noisy features by which beamforming and speech enhancement techniques are used, and additional speaker related features as well as  ... 
arXiv:1608.04983v1 fatcat:eilwhul6wvbnfe42iewna2y66u

Modulation Spectral Features for Robust Far-Field Speaker Identification

T.H. Falk, Wai-Yip Chan
2010 IEEE Transactions on Audio, Speech, and Language Processing  
In this paper, auditory inspired modulation spectral features are used to improve automatic speaker identification (ASI) performance in the presence of room reverberation.  ...  An eight-channel modulation filterbank is then applied to the temporal envelope of each gammatone filter output.  ...  Modulation Spectral Features for Robust Far-Field Speaker Identification Tiago H.  ... 
doi:10.1109/tasl.2009.2023679 fatcat:aonxticob5gwfbenxognjtzx4e

Multiple views of the response of an ensemble of spectro-temporal features support concurrent classification of utterance, prosody, sex and speaker identity

M. Coath, J. M. Brader, S. Fusi, S. L. Denham
2005 Network  
of the speaker.  ...  We also show that the responses of the ensemble are sparse in the sense that a small number of features respond for each stimulus type.  ...  Speaker identification results This was the only experiment that did not use the ISOLET corpus.  ... 
doi:10.1080/09548980500290120 pmid:16411500 fatcat:yhfuxebuhjbpbi4henhrrpyb5q

Employing both gender and emotion cues to enhance speaker identification performance in emotional talking environments

Ismail Mohd Adnan Shahin
2013 International Journal of Speech Technology  
The results of this work show that speaker identification performance based on using both gender and emotion cues is higher than that based on using gender cues only, emotion cues only, and neither gender  ...  The achieved average speaker identification performance based on the new proposed approach falls within 2.35% of that obtained in subjective evaluation by human judges.  ...  [14] aimed in one of their works to enhance the automatic emotional speech classification methods using ensemble or multi-classifier system (MCS) approaches.  ... 
doi:10.1007/s10772-013-9188-2 fatcat:bddyw57o7ndixo2mcnmisocrqq

Efficient band selection for improving the robustness of the EMD-based cepstral features

Ehsan Samadi, Ghasem Alipoor
2019 Sadhana (Bangalore)  
Simulation results show that, using the proposed features for automatic gender identification considerably improves the performance of the system, in particular in noisy environments. adhana(0123456789  ...  To address this problem, in the present study, we investigate the application of Empirical Mode Decomposition (EMD) in extracting more efficient and robust features for automatic gender identification.  ...  Furthermore, as a pre-processing unit, gender identification can enhance the accuracy of some recognition models, e.g., within the speaker identification [1] , speaker verification [2] and speaker diarization  ... 
doi:10.1007/s12046-019-1052-x fatcat:w4hsttz3yjaldenp6tha5mba74

A text-independent speaker verification model: A comparative analysis [article]

Rishi Charan, Manisha.A, Karthik.R, Rajesh Kumar M
2017 arXiv   pre-print
In this paper, we explore the various methods available in each block in the process of speaker recognition with the objective to identify best of techniques that could be used to get precise results.  ...  The most pressing challenge in the field of voice biometrics is selecting the most efficient technique of speaker recognition.  ...  Md Raibul et al [2] have already worked on speaker identification which uses cepstral features and PCA for classification.  ... 
arXiv:1712.00917v1 fatcat:r7wejspexvdbpotlfryqpcoaji
« Previous Showing results 1 — 15 out of 6,052 results