Filters








36 Hits in 4.2 sec

Ensemble Hierarchical Extreme Learning Machine for Speech Dereverberation

Tassadaq Hussain, Sabato Marco Siniscalchi, Hsiao-Lan Sharon Wang, Yu Tsao, Salerno Valerio Mario, Wen-Hung Liao
2019 IEEE Transactions on Cognitive and Developmental Systems  
Index Terms-Ensemble learning, hierarchical extreme learning machines (HELMs), highway extreme learning machine, residual extreme learning machine, speech dereverberation.  ...  In this article, we propose to address the data requirement issue while preserving the advantages of deep neural structures leveraging upon hierarchical extreme learning machines (HELMs), which are not  ...  The proposed ensemble learning framework for speech dereverberation is then described. A. Extreme Learning Machines 1) ELM Model: The ELM model was proposed by Huang et al.  ... 
doi:10.1109/tcds.2019.2953620 fatcat:rgfik77bdjeyvhdkmapl25py5m

2020 Index IEEE Transactions on Cognitive and Developmental Systems Vol. 12

2020 IEEE Transactions on Cognitive and Developmental Systems  
., +, TCDS Dec. 2020 797-808 Extreme learning machines Ensemble Hierarchical Extreme Learning Machine for Speech Dereverberation.  ...  Gong, P., +, TCDS March 2020 73-83 Speech enhancement Ensemble Hierarchical Extreme Learning Machine for Speech Dereverberation.  ... 
doi:10.1109/tcds.2020.3044690 fatcat:yfo6c366aramfdltqegqyqphbq

Supervised Speech Separation Based on Deep Learning: An Overview [article]

DeLiang Wang, Jitong Chen
2018 arXiv   pre-print
Then we discuss three main components of supervised separation: learning machines, training targets, and acoustic features.  ...  Much of the overview is on separation algorithms where we review monaural methods, including speech enhancement (speech-nonspeech separation), speaker separation (multi-talker separation), and speech dereverberation  ...  We thank Masood Delfarah for help in manuscript preparation and Jun Du, Yu Tsao, Yuxuan Wang, Yong Xu, and Xueliang Zhang for helpful comments on an earlier version.  ... 
arXiv:1708.07524v2 fatcat:bvaa2yuppffppnta2lfpkk4v4m

A Subband-Based SVM Front-End for Robust ASR [article]

Jibran Yousafzai and Zoran Cvetkovic and Peter Sollich and Matthew Ager
2013 arXiv   pre-print
This work proposes a novel support vector machine (SVM) based robust automatic speech recognition (ASR) front-end that operates on an ensemble of the subband components of high-dimensional acoustic waveforms  ...  The key issues of selecting the appropriate SVM kernels for classification in frequency subbands and the combination of individual subband classifiers using ensemble methods are addressed.  ...  Majority voting is the simplest uniform aggregation scheme commonly used in machine learning.  ... 
arXiv:1401.3322v1 fatcat:xjrhwhbiv5d5lgax5sfzbxjk6y

ICASSP 2020 Table of Contents

2020 ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)  
TOPOLOGY OPTIMIZATION FOR IMAGE DENOISING Wengtai Su, National Tsing Hua University, Taiwan; Gene Cheung, Richard P.  ...  IDSP-P1: EMERGING SIGNAL PROCESSING APPLICATIONS IDSP-P1.1: JOINT TRAINING OF DEEP NEURAL NETWORKS FOR MULTI-CHANNEL ........................................ 3032 DEREVERBERATION AND SPEECH SOURCE SEPARATION  ...  SPATIO-TEMPORAL ......................................... 2388 VIDEO ACTION DETECTION Yutang Wu, Hanli Wang, Shuheng Wang, Qinyu Li, Tongji University, China IVMSP-P6: MACHINE LEARNING FOR IMAGE/VIDEO  ... 
doi:10.1109/icassp40776.2020.9054406 fatcat:6h7hh2hxhne4pbmphharu2et2m

Deep Learning for Distant Speech Recognition [article]

Mirco Ravanelli
2017 arXiv   pre-print
Among the other achievements, building computers that understand speech represents a crucial leap towards intelligent machines.  ...  Deep learning is an emerging technology that is considered one of the most promising directions for reaching higher levels of artificial intelligence.  ...  These efforts were extremely important for the research community, since they laid the foundations for the basic learning algorithms [23] .  ... 
arXiv:1712.06086v1 fatcat:2b7ymqmihjan5nkxeqrxq52wki

Review: Deep Learning in Electron Microscopy [article]

Jeffrey M. Ede
2020 arXiv   pre-print
For context, we review popular applications of deep learning in electron microscopy.  ...  Finally, we discuss future directions of deep learning in electron microscopy.  ...  Acknowledgements Thanks go to Jeremy Sloan and Martin Lotz for internally reviewing this article.  ... 
arXiv:2009.08328v4 fatcat:umocfp5dgvfqzck4ontlflh5ca

An Overview of Noise-Robust Automatic Speech Recognition

Jinyu Li, Li Deng, Yifan Gong, Reinhold Haeb-Umbach
2014 IEEE/ACM Transactions on Audio Speech and Language Processing  
New waves of consumer-centric applications, such as voice search and voice interaction with mobile devices and home entertainment systems, increasingly require automatic speech recognition (ASR) to be  ...  To this end, it is critical to establish a solid, consistent, and common mathematical foundation for noise-robust ASR, which is lacking at present.  ...  ACKNOWLEDGMENT The authors want to thank Jon Grossmann for proofreading the manuscript.  ... 
doi:10.1109/taslp.2014.2304637 fatcat:xs3qq2bo3jd5toozh5bc55ae6u

Adsorption and diffusion on a stepped surface: Atomic hydrogen on Pt(211)

R. A. Olsen, Ş. C. Bădescu, S. C. Ying, E. J. Baerends
2004 Journal of Chemical Physics  
Additionally, some issues in machine learning techniques in prosody modeling will be discussed.  ...  Prosodic features are extracted to represent duration, pitch, and energy, with different normalization, and modeled using machine learning techniques.  ... 
doi:10.1063/1.1755664 pmid:15268219 fatcat:qgbl7hbggvan7eyuyiow2d4yhy

Auralization of an orchestra using multichannel and multisource technique

Michelle C. Vigeant, Lily M. Wang, Jens Holger Rindel
2006 Journal of the Acoustical Society of America  
Additionally, some issues in machine learning techniques in prosody modeling will be discussed.  ...  Prosodic features are extracted to represent duration, pitch, and energy, with different normalization, and modeled using machine learning techniques.  ...  These gliders periodically surfaced for GPS fixes and data transfer via satellite phone.  ... 
doi:10.1121/1.4787034 fatcat:brilcvuxpbgdvpdfzwovehoary

A transparency model and its applications for simulation of reflector arrays and sound transmission

Claus Lynge Christensen, Jens Holger Rindel
2006 Journal of the Acoustical Society of America  
Additionally, some issues in machine learning techniques in prosody modeling will be discussed.  ...  Prosodic features are extracted to represent duration, pitch, and energy, with different normalization, and modeled using machine learning techniques.  ...  These gliders periodically surfaced for GPS fixes and data transfer via satellite phone.  ... 
doi:10.1121/1.4786982 fatcat:uh2zucni3zaa5nnkivc5sh5tpq

Acoustics of ancient Greek and Roman theaters in use today

Anders Christian Gade, Konstantinos Angelakis
2006 Journal of the Acoustical Society of America  
Speech and noise levels in classrooms-signal, noise, reverberation levels as a metric for acoustic design for learning.  ...  These findings suggest that high variability training may direct attention to the accent-general properties of speech necessary for perceptual learning of accented speech.  ...  data available for the test site.  ... 
doi:10.1121/1.4787803 fatcat:3lczeegofreblpjmfixt7gdxv4

Effects of errorless learning on the acquisition of velopharyngeal movement control

Andus Wing-Kuen Wong, Tara Whitehill, Estella Ma, Rich Masters
2012 Journal of the Acoustical Society of America  
Computational skills allow rapid and automatic "statistical learning" and social interaction is necessary for this computational learning process to occur.  ...  Language learning and the developing brain: Cross-cultural studies unravel the effects of biology and culture.  ...  Normal speaking participants learned to produce hypernasal speech in either an errorless learning condition (in which the possibility for errors was limited) or an errorful learning condition (in which  ... 
doi:10.1121/1.4708235 fatcat:7wzupz5u2nd6nc7ttvbpxwvunm

Calculation of the A term of magnetic circular dichroism based on time dependent-density functional theory I. Formulation and implementation

Michael Seth, Tom Ziegler, Arup Banerjee, Jochen Autschbach, Stan J. A. van Gisbergen, Evert J. Baerends
2004 Journal of Chemical Physics  
In the extreme, minimizing reverberation times would lead to near anechoic rooms for speech and inadequate signal-to-noise ratios.  ...  The effect of noise on novel word learning in sequential bilingual children.  ...  Suppression of speech intelligibility loss through a modulation transfer function-based speech dereverberation method.  ... 
doi:10.1063/1.1747828 pmid:15268124 fatcat:hwyfald5a5f4njmw5h7p6vjhnu

Neural network modeling of a dolphin's sonar discrimination capabilities

Lars N. Andersen, A. René Rasmussen, Whitlow W. L. Au, Paul E. Nachtigall, Herbert Roitblat
1994 Journal of the Acoustical Society of America  
Processing of continuous speech by a hierarchical neural network.  ...  reading ability, as well as for success in learning a second language.  ...  Funds were used to purchase a DOS machine and a Macintosh, which run software including programs for signal processing, speech synthesis, statistics, word processing, and graphics.  ... 
doi:10.1121/1.410770 fatcat:ioiiov6bmjdi7kiflait5dhdfe
« Previous Showing results 1 — 15 out of 36 results