Latent Variable Analysis and Signal Separation

1. Front Matter

   Pages I-XVI

   PDF

2. Tensor-Based Methods for Blind Signal Separation

   1. Front Matter

      Pages 1-1

      PDF

   2. Stochastic and Deterministic Tensorization for Blind Signal Separation

      • Otto Debals, Lieven De Lathauwer

      Pages 3-13

   3. Block-Decoupling Multivariate Polynomials Using the Tensor Block-Term Decomposition

      • Philippe Dreesen, Thomas Goossens, Mariya Ishteva, Lieven De Lathauwer, Johan Schoukens

      Pages 14-21

   4. A Polynomial Formulation for Joint Decomposition of Symmetric Tensors of Different Orders

      • Pierre Comon, Yang Qi, Konstantin Usevich

      Pages 22-30

   5. Rank Splitting for CANDECOMP/PARAFAC

      • Anh-Huy Phan, Petr Tichavský, Andrzej Cichocki

      Pages 31-40

   6. Some Rank Conditions for the Identifiability of the Sparse Paralind Model

      • Sebastian Miron, David Brie

      Pages 41-48

   7. Tensors and Latent Variable Models

      • Mariya Ishteva

      Pages 49-55

   8. Tensor Factorisation Approach for Separation of Convolutive Complex Communication Signals

      • Samaneh Kouchaki, Saeid Sanei

      Pages 56-63

   9. Enhanced Tensor Based Semi-blind Estimation Algorithm for Relay-Assisted MIMO Systems

      • Jianshu Zhang, Ahmad Nimr, Kristina Naskovska, Martin Haardt

      Pages 64-72

3. Deep Neural Networks for Supervised Speech Separation/Enhancement

   1. Front Matter

      Pages 73-73

      PDF

   2. Improving Deep Neural Network Based Speech Enhancement in Low SNR Environments

      • Tian Gao, Jun Du, Yong Xu, Cong Liu, Li-Rong Dai, Chin-Hui Lee

      Pages 75-82

   3. Noise Perturbation Improves Supervised Speech Separation

      • Jitong Chen, Yuxuan Wang, DeLiang Wang

      Pages 83-90

   4. Speech Enhancement with LSTM Recurrent Neural Networks and its Application to Noise-Robust ASR

      • Felix Weninger, Hakan Erdogan, Shinji Watanabe, Emmanuel Vincent, Jonathan Le Roux, John R. Hershey et al.

      Pages 91-99

   5. Adaptive Denoising Autoencoders: A Fine-Tuning Scheme to Learn from Test Mixtures

      • Minje Kim, Paris Smaragdis

      Pages 100-107

4. Joint Analysis of Multiple Datasets, Data Fusion, and Related Topics

   1. Front Matter

      Pages 109-109

      PDF

   2. Joint Independent Subspace Analysis: A Quasi-Newton Algorithm

      • Dana Lahat, Christian Jutten

      Pages 111-118

   3. Joint Decompositions with Flexible Couplings

      • Rodrigo Cabral Farias, Jérémy Emile Cohen, Christian Jutten, Pierre Comon

      Pages 119-126

   4. Learning Coupled Embedding Using MultiView Diffusion Maps

      • Ofir Lindenbaum, Arie Yeredor, Moshe Salhov

      Pages 127-134

   5. Extraction of Temporal Patterns in Multi-rate and Multi-modal Datasets

      • Antoine Liutkus, Umut Şimşekli, A. Taylan Cemgil

      Pages 135-142

This book constitutes the proceedings of the 12th International Conference on Latent Variable Analysis and Signal Separation, LVA/ICS 2015, held in Liberec, Czech Republic, in August 2015. The 61 revised full papers presented – 29 accepted as oral presentations and 32 accepted as poster presentations – were carefully reviewed and selected from numerous submissions. Five special topics are addressed: tensor-based methods for blind signal separation; deep neural networks for supervised speech separation/enhancement; joined analysis of multiple datasets, data fusion, and related topics; advances in nonlinear blind source separation; sparse and low rank modeling for acoustic signal processing.

• audio signal processing
• augmented statistics
• classification
• computational imaging
• decision trees
• deep learning
• empirical kernel maps
• environmental monitoring
• independent vector analysis
• joint block diagonalization
• medical imaging
• mobile sensor network
• multilinear algebra
• real-time
• similarity measures
• sparse modeling
• speech separation
• statistical modeling
• supervised learning
• texture analysis

• Inria, Villers-les-Nancy, France

  Emmanuel Vincent

• Tel Aviv University, Tel-Aviv, Israel

  Arie Yeredor

• Technical University of Libere, Liberec, Czech Republic

  Zbyněk Koldovský

• The Czech Academy of Sciences, Prague, Czech Republic

  Petr Tichavský

Policies and ethics