Nonlinear Analyses and Algorithms for Speech Processing

1. Front Matter

   PDF

2. The COST-277 European Action: An Overview

   1. The COST-277 European Action: An Overview

      • Marcos Faundez-Zanuy, Unto Laine, Gernot Kubin, Stephen McLaughlin, Bastiaan Kleijn, Gerard Chollet et al.

      Pages 1-9

   2. Neuro-fuzzy Logic in Signal Processing for Communications: From Bits to Protocols

      • Ana Pérez-Neira, Miguel A. Lagunas, Antoni Morell, Joan Bas

      Pages 10-36

   3. Connected Operators for Signal and Image Processing

      • Philippe Salembier

      Pages 37-65

3. Speaker Recognition

   1. Exploiting High-Level Information Provided by ALISP in Speaker Recognition

      • Asmaa El Hannani, Dijana Petrovska-Delacrétaz

      Pages 66-71

   2. MLP Internal Representation as Discriminative Features for Improved Speaker Recognition

      • Dalei Wu, Andrew Morris, Jacques Koreman

      Pages 72-80

   3. Weighting Scores to Improve Speaker-Dependent Threshold Estimation in Text-Dependent Speaker Verification

      • Javier R. Saeta, Javier Hernando

      Pages 81-91

   4. Parameter Optimization in a Text-Dependent Cryptographic-Speech-Key Generation Task

      • L. Paola García-Perera, Juan A. Nolazco-Flores, Carlos Mex-Perera

      Pages 92-99

   5. The COST-277 Speech Database

      • Marcos Faundez-Zanuy, Martin Hagmüller, Gernot Kubin, W. Bastiaan Kleijn

      Pages 100-107

4. Speech Analysis

   1. Children’s Organization of Discourse Structure Through Pausing Means

      • Anna Esposito

      Pages 108-115

   2. F0 and Intensity Distributions of Marsec Speakers: Types of Speaker Prosody

      • Brigitte Zellner Keller

      Pages 116-124

   3. A Two-Level Drive – Response Model of Non-stationary Speech Signals

      • Friedhelm R. Drepper

      Pages 125-138

   4. Advanced Methods for Glottal Wave Extraction

      • Jacqueline Walker, Peter Murphy

      Pages 139-149

   5. Cepstrum-Based Estimation of the Harmonics-to-Noise Ratio for Synthesized and Human Voice Signals

      • Peter J. Murphy, Olatunji O. Akande

      Pages 150-160

   6. Pseudo Cepstral Analysis of Czech Vowels

      • Robert Vích

      Pages 161-173

   7. Bispectrum Estimators for Voice Activity Detection and Speech Recognition

      • J. M. Górriz, C. G. Puntonet, J. Ramírez, J. C. Segura

      Pages 174-185

5. Voice Pathologies

   1. On the Acoustic-to-Electropalatographic Mapping

      • Asterios Toutios, Konstantinos Margaritis

      Pages 186-195

   2. Issues in Clinical Applications of Bilateral Multi-step Predictive Analysis of Speech

      • J. Schoentgen, E. Dessalle, A. Kacha, F. Grenez

      Pages 196-205

   3. Optimal Size of Time Window in Nonlinear Features for Voice Quality Measurement

      • Jesús B. Alonso, Fernando Díaz-de-María, Carlos M. Travieso, Miguel A. Ferrer

      Pages 206-218

   4. Support Vector Machines Applied to the Detection of Voice Disorders

      • Juan Ignacio Godino-Llorente, Pedro Gómez-Vilda, Nicolás Sáenz-Lechón, Manuel Blanco-Velasco, Fernando Cruz-Roldán, Miguel Angel Ferrer-Ballester

      Pages 219-230

We present in this volume the collection of ?nally accepted papers of NOLISP 2005 conference. It has been the third event in a series of events related to N- linear speech processing, in the framework of the European COST action 277 “Nonlinear speech processing”. Many speci?cs of the speech signal are not well addressed by conv- tional models currently used in the ?eld of speech processing. The purpose of NOLISP is to present and discuss novel ideas, work and results related to alternative techniques for speech processing, which depart from mainstream approaches. With this intention in mind, we provide an open forum for discussion. Alt- nate approaches are appreciated, although the results achieved at present may not clearly surpass results based on state-of-the-art methods. The call for papers was launched at the beginning of 2005, addressing the following domains: 1. Non-Linear Approximation and Estimation 2. Non-Linear Oscillators and Predictors 3. Higher-Order Statistics 4. Independent Component Analysis 5. Nearest Neighbors 6. Neural Networks 7. Decision Trees 8. Non-Parametric Models 9. Dynamics of Non-Linear Systems 10. Fractal Methods 11. Chaos Modeling 12. Non-Linear Di?erential Equations 13. Others All the main ?elds of speech processing are targeted by the workshop, namely: 1. Speech Coding:Thebit rateavailablefor speechsignalsmustbe strictly l- ited in order to accommodate the constraints of the channel resource.

• ai
• algorithms
• artificial intelligence
• higher-order statistics
• image processing
• intelligent information processing
• modeling
• neural computation
• neuro-fuzzy logic
• non-linear approximation
• non-parametric models
• speaker recognition
• speech
• speech processing
• speech recognition

• Escola Universitària Politècnica de Mataró, UPC, Spain

  Marcos Faundez-Zanuy

• Escola Universitària Politècnica de Mataró, Spain

  Léonard Janer, Antonio Satue-Villar

• Department of Psychology, Second University of Naples, and IIASS, Vietri sul Mare, Italy

  Anna Esposito

• The Auton Lab, Carnegie Mellon University, Pittsburgh, USA

  Josep Roure

• Escola Universitària Politècnica de Mataró (UPC), Barcelona, Spain

  Virginia Espinosa-Duro

Policies and ethics