Sorry, you need to enable JavaScript to visit this website.

ICASSP 2020

ICASSP is the world’s largest and most comprehensive technical conference focused on signal processing and its applications. The ICASSP 2020 conference will feature world-class presentations by internationally renowned speakers, cutting-edge session topics and provide a fantastic opportunity to network with like-minded professionals from around the world. Visit website.

Mockingjay: Unsupervised Speech Representation Learning with Deep Bidirectional Transformer Encoders


We present Mockingjay as a new speech representation learning approach, where bidirectional Transformer encoders are pre-trained on a large amount of unlabeled speech. Previous speech representation methods learn through conditioning on past frames and predicting information about future frames. Whereas Mockingjay is designed to predict the current frame through jointly conditioning on both past and future contexts.

Paper Details

Authors:
Andy T. Liu, Shu-wen Yang, Po-Han Chi, Po-chun Hsu, Hung-yi Lee
Submitted On:
15 May 2020 - 10:18pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

Presentation Slides

(9)

Subscribe

[1] Andy T. Liu, Shu-wen Yang, Po-Han Chi, Po-chun Hsu, Hung-yi Lee, "Mockingjay: Unsupervised Speech Representation Learning with Deep Bidirectional Transformer Encoders", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5364. Accessed: Jun. 06, 2020.
@article{5364-20,
url = {http://sigport.org/5364},
author = {Andy T. Liu; Shu-wen Yang; Po-Han Chi; Po-chun Hsu; Hung-yi Lee },
publisher = {IEEE SigPort},
title = {Mockingjay: Unsupervised Speech Representation Learning with Deep Bidirectional Transformer Encoders},
year = {2020} }
TY - EJOUR
T1 - Mockingjay: Unsupervised Speech Representation Learning with Deep Bidirectional Transformer Encoders
AU - Andy T. Liu; Shu-wen Yang; Po-Han Chi; Po-chun Hsu; Hung-yi Lee
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5364
ER -
Andy T. Liu, Shu-wen Yang, Po-Han Chi, Po-chun Hsu, Hung-yi Lee. (2020). Mockingjay: Unsupervised Speech Representation Learning with Deep Bidirectional Transformer Encoders. IEEE SigPort. http://sigport.org/5364
Andy T. Liu, Shu-wen Yang, Po-Han Chi, Po-chun Hsu, Hung-yi Lee, 2020. Mockingjay: Unsupervised Speech Representation Learning with Deep Bidirectional Transformer Encoders. Available at: http://sigport.org/5364.
Andy T. Liu, Shu-wen Yang, Po-Han Chi, Po-chun Hsu, Hung-yi Lee. (2020). "Mockingjay: Unsupervised Speech Representation Learning with Deep Bidirectional Transformer Encoders." Web.
1. Andy T. Liu, Shu-wen Yang, Po-Han Chi, Po-chun Hsu, Hung-yi Lee. Mockingjay: Unsupervised Speech Representation Learning with Deep Bidirectional Transformer Encoders [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5364

A NOVEL RANK SELECTION SCHEME IN TENSOR RING DECOMPOSITION BASED ON REINFORCEMENT LEARNING FOR DEEP NEURAL NETWORKS


Tensor decomposition has been proved to be effective for solving many problems in signal processing and machine learning. Recently, tensor decomposition finds its advantage for compressing deep neural networks. In many applications of deep neural networks, it is critical to reduce the number of parameters and computation workload to accelerate inference speed in deployment of the network. Modern deep neural network consists of multiple layers with multi-array weights where tensor decomposition is a natural way to perform compression.

Paper Details

Authors:
Submitted On:
15 May 2020 - 10:03pm
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

ICASSP_slides_Cheng.pdf

(10)

Subscribe

[1] , "A NOVEL RANK SELECTION SCHEME IN TENSOR RING DECOMPOSITION BASED ON REINFORCEMENT LEARNING FOR DEEP NEURAL NETWORKS", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5363. Accessed: Jun. 06, 2020.
@article{5363-20,
url = {http://sigport.org/5363},
author = { },
publisher = {IEEE SigPort},
title = {A NOVEL RANK SELECTION SCHEME IN TENSOR RING DECOMPOSITION BASED ON REINFORCEMENT LEARNING FOR DEEP NEURAL NETWORKS},
year = {2020} }
TY - EJOUR
T1 - A NOVEL RANK SELECTION SCHEME IN TENSOR RING DECOMPOSITION BASED ON REINFORCEMENT LEARNING FOR DEEP NEURAL NETWORKS
AU -
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5363
ER -
. (2020). A NOVEL RANK SELECTION SCHEME IN TENSOR RING DECOMPOSITION BASED ON REINFORCEMENT LEARNING FOR DEEP NEURAL NETWORKS. IEEE SigPort. http://sigport.org/5363
, 2020. A NOVEL RANK SELECTION SCHEME IN TENSOR RING DECOMPOSITION BASED ON REINFORCEMENT LEARNING FOR DEEP NEURAL NETWORKS. Available at: http://sigport.org/5363.
. (2020). "A NOVEL RANK SELECTION SCHEME IN TENSOR RING DECOMPOSITION BASED ON REINFORCEMENT LEARNING FOR DEEP NEURAL NETWORKS." Web.
1. . A NOVEL RANK SELECTION SCHEME IN TENSOR RING DECOMPOSITION BASED ON REINFORCEMENT LEARNING FOR DEEP NEURAL NETWORKS [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5363

Look globally, age locally: Face aging with an attention mechanism


Face aging is of great importance for cross-age recognition and entertainment-related applications. Recently, conditional generative adversarial networks (cGANs) have achieved impressive results for facial aging. Existing cGANs-based methods usually require a pixel-wise loss to keep the identity and background consistent. However, minimizing the pixel-wise loss between the input and synthesized images likely resulting in a ghosted or blurry face.

Paper Details

Authors:
Submitted On:
15 May 2020 - 9:17pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

slide_paper5430.pdf

(13)

Subscribe

[1] , "Look globally, age locally: Face aging with an attention mechanism", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5362. Accessed: Jun. 06, 2020.
@article{5362-20,
url = {http://sigport.org/5362},
author = { },
publisher = {IEEE SigPort},
title = {Look globally, age locally: Face aging with an attention mechanism},
year = {2020} }
TY - EJOUR
T1 - Look globally, age locally: Face aging with an attention mechanism
AU -
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5362
ER -
. (2020). Look globally, age locally: Face aging with an attention mechanism. IEEE SigPort. http://sigport.org/5362
, 2020. Look globally, age locally: Face aging with an attention mechanism. Available at: http://sigport.org/5362.
. (2020). "Look globally, age locally: Face aging with an attention mechanism." Web.
1. . Look globally, age locally: Face aging with an attention mechanism [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5362

Graph Neural Net using Analytical Graph Filters and Topology Optimization for Image Denoising


While convolutional neural nets (CNN) have achieved remarkable performance for a wide range of inverse imaging applications, the filter coefficients are computed in a purely data-driven manner and are not explainable. Inspired by an analytically derived CNN by Hadji et al., in this paper we construct a new layered graph convolutional neural net (GCNN) using GraphBio as our graph filter.

Paper Details

Authors:
, Gene Cheung, Richard Wildes and Chia-Wen Lin
Submitted On:
15 May 2020 - 8:51pm
Short Link:
Type:
Event:
Document Year:
Cite

Document Files

Icassp_DeepAGF_2020_slides

(12)

Subscribe

[1] , Gene Cheung, Richard Wildes and Chia-Wen Lin, "Graph Neural Net using Analytical Graph Filters and Topology Optimization for Image Denoising", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5361. Accessed: Jun. 06, 2020.
@article{5361-20,
url = {http://sigport.org/5361},
author = {; Gene Cheung; Richard Wildes and Chia-Wen Lin },
publisher = {IEEE SigPort},
title = {Graph Neural Net using Analytical Graph Filters and Topology Optimization for Image Denoising},
year = {2020} }
TY - EJOUR
T1 - Graph Neural Net using Analytical Graph Filters and Topology Optimization for Image Denoising
AU - ; Gene Cheung; Richard Wildes and Chia-Wen Lin
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5361
ER -
, Gene Cheung, Richard Wildes and Chia-Wen Lin. (2020). Graph Neural Net using Analytical Graph Filters and Topology Optimization for Image Denoising. IEEE SigPort. http://sigport.org/5361
, Gene Cheung, Richard Wildes and Chia-Wen Lin, 2020. Graph Neural Net using Analytical Graph Filters and Topology Optimization for Image Denoising. Available at: http://sigport.org/5361.
, Gene Cheung, Richard Wildes and Chia-Wen Lin. (2020). "Graph Neural Net using Analytical Graph Filters and Topology Optimization for Image Denoising." Web.
1. , Gene Cheung, Richard Wildes and Chia-Wen Lin. Graph Neural Net using Analytical Graph Filters and Topology Optimization for Image Denoising [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5361

NORMALIZED LEAST-MEAN-SQUARE ALGORITHMS WITH MINIMAX CONCAVE PENALTY


We propose a novel problem formulation for sparsity-aware adaptive filtering based on the nonconvex minimax concave (MC) penalty, aiming to obtain a sparse solution with small estimation bias. We present two algorithms: the first algorithm uses a single firm-shrinkage operation, while the second one uses double soft-shrinkage operations. The twin soft-shrinkage operations compensate each other, promoting sparsity while avoiding a serious increase of biases. The whole cost function is convex in certain parameter settings, while the instantaneous cost function is always nonconvex.

Paper Details

Authors:
Hiroyuki Kaneko, Masahiro Yukawa
Submitted On:
15 May 2020 - 8:00pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

icassp2020_slide_final.pdf

(10)

Subscribe

[1] Hiroyuki Kaneko, Masahiro Yukawa, "NORMALIZED LEAST-MEAN-SQUARE ALGORITHMS WITH MINIMAX CONCAVE PENALTY", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5360. Accessed: Jun. 06, 2020.
@article{5360-20,
url = {http://sigport.org/5360},
author = {Hiroyuki Kaneko; Masahiro Yukawa },
publisher = {IEEE SigPort},
title = {NORMALIZED LEAST-MEAN-SQUARE ALGORITHMS WITH MINIMAX CONCAVE PENALTY},
year = {2020} }
TY - EJOUR
T1 - NORMALIZED LEAST-MEAN-SQUARE ALGORITHMS WITH MINIMAX CONCAVE PENALTY
AU - Hiroyuki Kaneko; Masahiro Yukawa
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5360
ER -
Hiroyuki Kaneko, Masahiro Yukawa. (2020). NORMALIZED LEAST-MEAN-SQUARE ALGORITHMS WITH MINIMAX CONCAVE PENALTY. IEEE SigPort. http://sigport.org/5360
Hiroyuki Kaneko, Masahiro Yukawa, 2020. NORMALIZED LEAST-MEAN-SQUARE ALGORITHMS WITH MINIMAX CONCAVE PENALTY. Available at: http://sigport.org/5360.
Hiroyuki Kaneko, Masahiro Yukawa. (2020). "NORMALIZED LEAST-MEAN-SQUARE ALGORITHMS WITH MINIMAX CONCAVE PENALTY." Web.
1. Hiroyuki Kaneko, Masahiro Yukawa. NORMALIZED LEAST-MEAN-SQUARE ALGORITHMS WITH MINIMAX CONCAVE PENALTY [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5360

Mr Nikolajs Skuratovs


In this paper we consider the problem of recovering a signal x of size N from noisy and compressed measurements y = A x + w of size M, where the measurement matrix A is right-orthogonally invariant (ROI). Vector Approximate Message Passing (VAMP) demonstrates great reconstruction results for even highly ill-conditioned matrices A in relatively few iterations. However, performing each iteration is challenging due to either computational or memory point of view.

Paper Details

Authors:
Nikolajs Skuratovs, Michael Davies
Submitted On:
15 May 2020 - 7:04pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP_presentation_v2_pdf.pdf

(9)

Subscribe

[1] Nikolajs Skuratovs, Michael Davies, "Mr Nikolajs Skuratovs", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5359. Accessed: Jun. 06, 2020.
@article{5359-20,
url = {http://sigport.org/5359},
author = {Nikolajs Skuratovs; Michael Davies },
publisher = {IEEE SigPort},
title = {Mr Nikolajs Skuratovs},
year = {2020} }
TY - EJOUR
T1 - Mr Nikolajs Skuratovs
AU - Nikolajs Skuratovs; Michael Davies
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5359
ER -
Nikolajs Skuratovs, Michael Davies. (2020). Mr Nikolajs Skuratovs. IEEE SigPort. http://sigport.org/5359
Nikolajs Skuratovs, Michael Davies, 2020. Mr Nikolajs Skuratovs. Available at: http://sigport.org/5359.
Nikolajs Skuratovs, Michael Davies. (2020). "Mr Nikolajs Skuratovs." Web.
1. Nikolajs Skuratovs, Michael Davies. Mr Nikolajs Skuratovs [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5359

Frame-based Overlapping Speech Detection using Convolutional Neural Networks


Naturalistic speech recordings usually contain speech signals from multiple speakers. This phenomenon can degrade the performance of speech technologies due to the complexity of tracing and recognizing individual speakers. In this study, we investigate the detection of overlapping speech on segments as short as 25 ms using Convolutional Neural Networks. We evaluate the detection performance using different spectral features, and show that pyknogram features outperforms other commonly used speech features.

Paper Details

Authors:
Midia Yousefi, Hohn H.L. Hansen
Submitted On:
15 May 2020 - 6:57pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

ICASSP2020-overlap-detection_MY-JH-Mar30-2020.pdf

(8)

Subscribe

[1] Midia Yousefi, Hohn H.L. Hansen, "Frame-based Overlapping Speech Detection using Convolutional Neural Networks", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5358. Accessed: Jun. 06, 2020.
@article{5358-20,
url = {http://sigport.org/5358},
author = {Midia Yousefi; Hohn H.L. Hansen },
publisher = {IEEE SigPort},
title = {Frame-based Overlapping Speech Detection using Convolutional Neural Networks},
year = {2020} }
TY - EJOUR
T1 - Frame-based Overlapping Speech Detection using Convolutional Neural Networks
AU - Midia Yousefi; Hohn H.L. Hansen
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5358
ER -
Midia Yousefi, Hohn H.L. Hansen. (2020). Frame-based Overlapping Speech Detection using Convolutional Neural Networks. IEEE SigPort. http://sigport.org/5358
Midia Yousefi, Hohn H.L. Hansen, 2020. Frame-based Overlapping Speech Detection using Convolutional Neural Networks. Available at: http://sigport.org/5358.
Midia Yousefi, Hohn H.L. Hansen. (2020). "Frame-based Overlapping Speech Detection using Convolutional Neural Networks." Web.
1. Midia Yousefi, Hohn H.L. Hansen. Frame-based Overlapping Speech Detection using Convolutional Neural Networks [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5358

LOW MUTUAL AND AVERAGE COHERENCE DICTIONARY LEARNING USING CONVEX APPROXIMATION

Paper Details

Authors:
Submitted On:
15 May 2020 - 6:11pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

1866.pdf

(8)

Subscribe

[1] , "LOW MUTUAL AND AVERAGE COHERENCE DICTIONARY LEARNING USING CONVEX APPROXIMATION", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5357. Accessed: Jun. 06, 2020.
@article{5357-20,
url = {http://sigport.org/5357},
author = { },
publisher = {IEEE SigPort},
title = {LOW MUTUAL AND AVERAGE COHERENCE DICTIONARY LEARNING USING CONVEX APPROXIMATION},
year = {2020} }
TY - EJOUR
T1 - LOW MUTUAL AND AVERAGE COHERENCE DICTIONARY LEARNING USING CONVEX APPROXIMATION
AU -
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5357
ER -
. (2020). LOW MUTUAL AND AVERAGE COHERENCE DICTIONARY LEARNING USING CONVEX APPROXIMATION. IEEE SigPort. http://sigport.org/5357
, 2020. LOW MUTUAL AND AVERAGE COHERENCE DICTIONARY LEARNING USING CONVEX APPROXIMATION. Available at: http://sigport.org/5357.
. (2020). "LOW MUTUAL AND AVERAGE COHERENCE DICTIONARY LEARNING USING CONVEX APPROXIMATION." Web.
1. . LOW MUTUAL AND AVERAGE COHERENCE DICTIONARY LEARNING USING CONVEX APPROXIMATION [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5357

Adaptive prediction of financial time-series for decision-making using a tensorial aggregation approach


Economic and financial decision-making may cause a significant impact on government, society, and industries. Due to the increasing volume of data, decision science has become an interdisciplinary field of study, supported by efficient methods and models of data analysis. Our contributions lie exactly in the intersection of signal processing, tensorial algebra, and decision science. More precisely, we introduce a novel approach in which the data taken into account in the decision process is modeled as a tensor.

Paper Details

Authors:
Betania S.C. Campello, Leonardo T. Duarte, João M. T. Romano
Submitted On:
15 May 2020 - 5:45pm
Short Link:
Type:
Event:
Presenter's Name:
Paper Code:
Document Year:
Cite

Document Files

campello_icassp_final.pdf

(9)

Subscribe

[1] Betania S.C. Campello, Leonardo T. Duarte, João M. T. Romano, "Adaptive prediction of financial time-series for decision-making using a tensorial aggregation approach", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5356. Accessed: Jun. 06, 2020.
@article{5356-20,
url = {http://sigport.org/5356},
author = {Betania S.C. Campello; Leonardo T. Duarte; João M. T. Romano },
publisher = {IEEE SigPort},
title = {Adaptive prediction of financial time-series for decision-making using a tensorial aggregation approach},
year = {2020} }
TY - EJOUR
T1 - Adaptive prediction of financial time-series for decision-making using a tensorial aggregation approach
AU - Betania S.C. Campello; Leonardo T. Duarte; João M. T. Romano
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5356
ER -
Betania S.C. Campello, Leonardo T. Duarte, João M. T. Romano. (2020). Adaptive prediction of financial time-series for decision-making using a tensorial aggregation approach. IEEE SigPort. http://sigport.org/5356
Betania S.C. Campello, Leonardo T. Duarte, João M. T. Romano, 2020. Adaptive prediction of financial time-series for decision-making using a tensorial aggregation approach. Available at: http://sigport.org/5356.
Betania S.C. Campello, Leonardo T. Duarte, João M. T. Romano. (2020). "Adaptive prediction of financial time-series for decision-making using a tensorial aggregation approach." Web.
1. Betania S.C. Campello, Leonardo T. Duarte, João M. T. Romano. Adaptive prediction of financial time-series for decision-making using a tensorial aggregation approach [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5356

INSTANCE-BASED MODEL ADAPTATION FOR DIRECT SPEECH TRANSLATION

Paper Details

Authors:
Viet-Nhat Nguyen,Matteo Negri,Marco Turchi
Submitted On:
15 May 2020 - 4:49pm
Short Link:
Type:
Event:
Presenter's Name:
Document Year:
Cite

Document Files

Instance-Based Model Adaptation _For Direct Speech Translation.pdf

(5)

Subscribe

[1] Viet-Nhat Nguyen,Matteo Negri,Marco Turchi, "INSTANCE-BASED MODEL ADAPTATION FOR DIRECT SPEECH TRANSLATION", IEEE SigPort, 2020. [Online]. Available: http://sigport.org/5355. Accessed: Jun. 06, 2020.
@article{5355-20,
url = {http://sigport.org/5355},
author = {Viet-Nhat Nguyen;Matteo Negri;Marco Turchi },
publisher = {IEEE SigPort},
title = {INSTANCE-BASED MODEL ADAPTATION FOR DIRECT SPEECH TRANSLATION},
year = {2020} }
TY - EJOUR
T1 - INSTANCE-BASED MODEL ADAPTATION FOR DIRECT SPEECH TRANSLATION
AU - Viet-Nhat Nguyen;Matteo Negri;Marco Turchi
PY - 2020
PB - IEEE SigPort
UR - http://sigport.org/5355
ER -
Viet-Nhat Nguyen,Matteo Negri,Marco Turchi. (2020). INSTANCE-BASED MODEL ADAPTATION FOR DIRECT SPEECH TRANSLATION. IEEE SigPort. http://sigport.org/5355
Viet-Nhat Nguyen,Matteo Negri,Marco Turchi, 2020. INSTANCE-BASED MODEL ADAPTATION FOR DIRECT SPEECH TRANSLATION. Available at: http://sigport.org/5355.
Viet-Nhat Nguyen,Matteo Negri,Marco Turchi. (2020). "INSTANCE-BASED MODEL ADAPTATION FOR DIRECT SPEECH TRANSLATION." Web.
1. Viet-Nhat Nguyen,Matteo Negri,Marco Turchi. INSTANCE-BASED MODEL ADAPTATION FOR DIRECT SPEECH TRANSLATION [Internet]. IEEE SigPort; 2020. Available from : http://sigport.org/5355

Pages