39th AES Conference - Program

INTRODUCTION PROGRAM EXHIBITORS VENUE REGISTRATION CONTACT

  

Paper Abstracts as PDF

 

Thursday, June 17th 2010

 

08.30

Registration is open

10.00 - 10.15

Opening

10.15 - 12.00

Exhibitors presentations
Cedar Audio, UK
National Center of Forensic Media, University of Colorado, Denver, USA
DPA Microphones, Denmark

12.00 - 13.30

Lunch

13.00

Exhibitions open

 

Session: Authentication

13.30 - 14.00

Detecting butt-spliced edits in forensic digital audio recordings

Alan Cooper, Metropolitan Police, United Kingdom

14.00 - 14.30

Visualisation of magnetic features on analogue audiotapes is still an important task

Dagmar Boss, Bayerisches Landeskriminalamt, Germany

14.30 - 15.00

Statistical Tools for Multimedia Forensics: Compression Effects Analysis

Catalin Grigoras, Forensic Science Center, Romania

15.00 - 15.30

Coffee

 

Session: Speech and Forensics - Voice Identification

15.30 - 16.00

Digitally Disguised Voices

Eddy B. Brixen, EBB-consult, Denmark

16.00 - 16.30

Characterising Formant Tracks in Viennese Diphthongs for Forensic Speaker Comparison

Ewald Enzinger, Acoustics Research Institute, Austrian Academy of Sciences, Austria

16.30 - 17.00

Channel compensation for forensic speaker identification using inverse processing

Andrey Barinov, Speech Technology Center Ltd., Russian Federation
Sergey Koval,
Speech Technology Center Ltd., Russian Federation
Michael Stolbov,
Speech Technology Center Ltd., Russian Federation
Pavel Ignatov,
Speech Technology Center Ltd. , Russian Federation

17.00 - 17.30

Importance of the relative delay of glottal source harmonics in speaker identification and voice register classification

Aníbal Ferreira, University of Porto, Portugal
Ricardo Sousa,
University of Porto, Portugal

18.00 - 20.00

Dinner

20.00 - 21.30

 Workshop: Frontiers of Forensic Audio Investigation
Chairman: Gordon Reid, Managing Director, CEDAR Audio Ltd, UK

ISO17025 Accreditation in Forensic Audio
Anna Czajkowski, Control Risks, UK

Audio winnowing: Finding whom or what you want to hear from a recording
Anil Alexander, Head of R&D, GriffComm Ltd, Oxford, UK

Time, tide and technological changes wait for no person
Alan French, Audio and Video Forensics Manager, FTS, UK

21.30 -

The bar is open

 

Friday, June 18th 2010

 

07.00 - 08.30

Buffet breakfast

08.30 - 10.00

Tutorial : The likelihood-ratio framework for the evaluation of forensic-comparison evidence

Geoffrey Stewart Morrison 1,2

1 School of Language Studies, Australian National University, Canberra, Australia

2 School of Electrical Engineering & Telecommunications, University of New South Wales, Sydney, Australia

This tutorial provides an introduction to the forensic comparison of audio recordings in the likelihood-ratio framework. Examples are drawn from audio recordings of human voices, but the principles and techniques can be applied to audio recordings of any source. The tutorial covers the topics essential for an understanding of the likelihood-ratio framework and its application to the forensic comparison of audio recordings including: What is a forensic likelihood ratio? Why is the likelihood-ratio framework the logically correct way to evaluate forensic evidence? How is a forensic likelihood ratio calculated? How is the validity and reliability of the likelihood-ratio output of a forensic-comparison system evaluated? What factors affect the validity and reliability of a forensic comparison system and how might validity and reliability be improved?

10.00 - 10.30

Coffee (and exhibition)

10.30 - 11.00

Session: Enhancement of Noisy Recordings : Corrupted speech intelligibility improvement using an adaptive filtering based algorithm

Damian Ellwart, Gdansk University of Technology, Poland
Andrzej Czyzewski, Gdansk University of Technology, Poland 

11.00 - 11.30

Adjusting a commercial speech enhancement system to optimize intelligibility

Gaston Hilkhuysen, University College London. United Kingdom
Mark Huckvale, University College London, United Kingdom

11.30 - 12.00

Speech Enhancement by Adaptive Noise Cancellation: Problems, Algorithms and Limits

Joerg Bitzer, Fachhochschule Wilhelmshaven/Oldenburg/Elsfleth, Germany
Matthias Brandt, Fachhochschule Wilhelmshaven/Oldenburg/Elsfleth, Germany

12.00 - 13.30

Buffet Lunch

 

Session: Acoustical Forensics

13.30 - 14.00

Performance of closed-form acoustic scene decomposition for forensic analysis

Banu Gunel, University of Surrey, United Kingdom
Grigorios Nikolopoulos, University of Surrey, United Kingdom
Huseyin Hacihabiboglu, University of Surrey, United Kingdom
Ahmet Kondoz, University of Surrey, United Kingdom

14.00 - 14.30

Directional aspects of forensic gunshot recordings

Robert Maher, Montana State University, United States

 

Session: Laboratory Procedures

14.30 - 15.00

The Modern Forensic Audio Laboratory - a public sector perspective

Robin How, Metropolitan Police, United Kingdom

15.00 - 15.30

Computer Forensics for the Forensic Audio Professional

Jeff Smith, National Center for Media Forensics, University of Colorado Denver, United States
Marcus Rogers, Purdue University Cyber Forensics Lab Dept. of Computer & Information Technology / CERIAS, United States

15.30 - 15.45

Coffee (and exhibition)

15.45 - 17.15

Workshop: Overview of Digital Audio Authentication

Chairman: Catalin Grigoras, Forensic Science Center, Romania

This session will emphasize some of the latest developments in forensic authentication of digital recordings, like Electric Network Frequency (ENF) and lossy compression analysis. It will also present the most common source or errors, the nowadays technological limits and the need for quality assurance.

The presentations will be interactive, in a lecture-discussion format with images, case studies and relevant research results.

 

17.15 - 19.45

Excursion to Frederiksborg Castle

20.00

Banquet

 

Saturday, June 19th 2010

 

07.00 - 08.30

Buffet breakfast

08.30 - 10.00

Tutorial : SWGDE and Forensic Audio Standards

Michael Piper1, 4, David Hallimore2, 4, and John Powell3, 4

1 U.S. Secret Service, Washington, D.C., USA

2 Houston Police Department, Houston, TX, USA

3 Los Angeles Sheriff’s Department, USA

4 Scientific Working Group for Digital Evidence, USA

The Scientific Working Group on Digital Evidence (SWGDE) recently published a position paper on the U.S. National Research Council’s February 18, 2009 report to Congress entitled “Strengthening Forensic Science in the United States: A Path Forward” in which the Council set forth a broad overview of the state of forensic science in the United States along with several recommendations for improvement. An overview of The Report, SWGDE’s response, and its Audio Committee’s current projects will be presented.

In each of the topics we intend to touch on (Forensic Audio Best Practices, minimum standards, accreditation/certification, training, ENF, etc.), there will be many points suitable for audience input and discussion. One of the key purposes of this presentation is to solicit input from the forensic audio community for the work we are undertaking at SWGDE.

 

10.00 - 10.30

Coffee (and exhibition)

 

Session: Speech and Forensics - Automated Systems

10.30 - 11.00

Voice carving in Police Dialogue: Forensic application of Automatic Speaker Segmentation

Anil Alexander, Griff Comm Ltd, United Kingdom
Oscar Forth, Griff Comm Ltd, United Kingdom
Robin How, Metropolitan Police Forensic Audio Laboratory, United Kingdom

11.00- 11.30

Automatic Forensic Voice Comparison Using Recording Adapted Background Models

Timo Becker, Federal Criminal Police Office, Germany
Michael Jessen, Federal Criminal Police Office, Germany
Sebastian Alsbach, University of Applied Science Koblenz, Germany
Franz Broß, University of Applied Science Koblenz, Germany

 

Session: Speech quality and intelligibility assessment - 1

11.30 - 12.00

Bayesian Adaptive Method for Estimating Speech Intelligibility in Noise

Nikolay Gaubitch, Imperial College London, United Kingdom
Mike Brookes, Imperial College London, United Kingdom
Patrick Naylor, Imperial College London, United Kingdom

12.00 - 13.30

Lunch

 

Session: Speech quality and intelligibility assessment - 2

13.30 - 14.00

C-QUAL - A Validitation of PESQ Using Degradations Encountered in Forensic and Law Enforcement Audio

Dushyant Sharma, Imperial College London, United Kingdom
Gaston Hikhuysen, University College London, United Kingdom
Patrick Naylor, Imperial College London, United Kingdom

14.00 - 14.30

Objective measures of speech intelligibility in forensic applications

Andrea Paoloni, Fondazione Ugo Bordoni, Italy
Giovanni Costantini, Università Tor Vergata, Italy
Massimiliano Todisco, Università Tor Vergata, Italy

14.30 - 15.00

Measuring the Effect of Noise Reduction on Listening Effort

Mark Huckvale, University College London, United Kingdom
Ngawang Frasi, University College London, United Kingdom

15.00 - 15.30

 Intelligibility Testing as an Engineering Tool

Robert Fellows, HMGCC, United Kingdom
Kenneth Worrall, HMGCC , United Kingdom

15.30 - 16.00

Closing of conference

16.00

Coffee

 Conference sponsor:

EBB Consult

 
Facebook   Twitter   LinkedIn   Google+   YouTube   RSS News Feeds  
AES - Audio Engineering Society