William T. McQuaide Managing Editor Gerri M. Calamusa Senior Editor Mary Ellen Ilich Associate Editor

Transcription

1 AES REGIONAL OFFICES Europe Conventions Kerkstraat 122/1, BE 1653 Dworp, Belgium, Tel: , Fax: , for convention information: euroconventions@aes.org Europe Services B.P , FR Bry Sur Marne Cedex, France, Tel: , for membership and publication sales: euroservices@aes.org United Kingdom British Section, Audio Engineering Society Ltd., P. O. Box 645, Slough, SL1 8BJ UK, Tel: , Fax: , UK@aes.org Japan AES Japan Section, Yoyogi, Room 703, Shibuyaku-ku, Tokyo , Japan, Tel: , Fax: , aes_japan_section@aes.org AES REGIONS AND SECTIONS Eastern Region, USA/Canada Sections: Atlanta, Boston, District of Columbia, New York, Philadelphia, Toronto Student Sections: American University, Appalachian State University, Atlanta, Berklee College of Music, Boston University CDIA, Carnegie Mellon University, Duquesne University, Emerson College, Fredonia, Full Sail Real World Education, Hampton University, IADT Tampa, Institute of Audio Research, Ithaca College, London Ontario, McGill University, New England Institute of Art, New England School of Communications, New York University, Peabody Institute of Johns Hopkins University, Pennsylvania State University, Shenandoah University, University of Hartford, University of Massachusetts-Lowell, University of Miami, University of North Carolina at Asheville, William Paterson University, Worcester Polytechnic Institute Central Region, USA/Canada Sections: Central Indiana, Central Texas, Chicago, Cincinnati, Detroit, Kansas City, Nashville, Heartland, New Orleans, St. Louis, Upper Midwest, West Michigan Student Sections: Art Institute of Tennessee-Nashville, Ball State University, Belmont University, Capital University, Columbia College, Flashpoint Academy, Indiana University, Institute of Production & Recording, Kansas City Kansas Community College, Michigan Technological University, McNally Smith College of Music, Middle Tennessee State University, Purdue University, SAE Nashville, Ridgewater College, Hutchinson Campus, University of Arkansas-Pine Bluff, University of Central Missouri, University of Illinois, Urbana-Champaign, University of Memphis, University of Michigan, Webster University Western Region, USA/Canada Sections: Alberta, Colorado, Los Angeles, Pacific Northwest, Portland, Sacramento Valley, San Francisco, Southern Nevada, Vancouver Student Sections: American River Regional, Brigham Young University, California State University Chico, Cal Poly San Luis Obispo, Citrus College, Cogswell Polytechnical College, Conservatory of Recording Arts and Sciences, Ex pression College for Digital Arts, Long Beach City College, San Francisco State University, Stanford University, The Art Institute of Las Vegas, The Art Institute of Seattle, University of Colorado at Denver, University of Southern California Northern Region, Europe Sections: Belgian, British, Danish, Finnish, Moscow, Netherlands, Norwegian, St. Petersburg, Swedish Student Sections: All-Russian State Institute of Cinematography, Danish, London UK, Netherlands, Russian Academy of Music, St. Petersburg, University of Lulea-Pitea, University of York Central Region, Europe Sections: Austrian, Belarus, Czech, Central German, North German, South German, Hungarian, Lithuanian, Polish, Slovak, Swiss, Ukrainian Student Sections: Aachen, Berlin, Czech Republic, Darmstadt, Detmold, Düsseldorf, Graz, Hamburg, Krakow, Ilmenau, Technical University of Gdansk (Poland), Vienna, Wroclaw University of Technology Southern Region, Europe Sections: Bulgarian, Croatian, French, Greek, Israel, Italian, Portugal, Romanian, Spanish, Serbia, Slovenia, Turkish Student Sections: Croatian, Conservatoire de Paris, Galatasaray ITM, Istanbul Bilgi University, Louis-Lumière, MicroFusa Barcelona, Serbian Latin American Region Sections: Argentina, Brazil, Chile, Colombia, Guatemala, Litoral, Mexico, Peru, Puebla, Rosario, Uruguay Student Sections: Academia de Musica Fermatta, Cochabamba, Del Bosque University, ECOS Escuela de Sonido, Escuela de Musica y Audio Fernando Sor, G Martell, I.A.V., Javeriana University, Instituto Mendocino de Audio y Sonido, Los Andes University, Orson Welles Institute, ORT Institute, Sala de Audio, Sonar Escuela de Sonido International Region Sections: Adelaide, Brisbane, Hong Kong, Japan, Korea, Malaysia, Melbourne, Philippines, Singapore, Sydney Student Section: Japan Durand Begault spatial perception and processing Eddy Bøgh Brixen audio forensics Elizabeth Cohen audio archiving, storage, and restoration Jeremy Cooperstock audio networking Diemer de Vries room acoustics and architectural acoustics Christof Faller low bit-rate audio coding Laurence R. Fincham instrumentation and measurement William T. McQuaide Managing Editor Gerri M. Calamusa Senior Editor Mary Ellen Ilich Associate Editor EDITORIAL STAFF John Vanderkooy Editor Associate Technical Editors Tomlinson Holman multichannel sound Stanley P. Lipshitz loudspeakers and sound reinforcement systems Robert C. Maher analysis and synthesis of sound Jan A. Pedersen signal processing Robert B. Schulein microphones Thomas Sporer psychoacoustics, perception, and listening tests Ingeborg M. Stochmal Copy Editor Barry A. Blesser Consulting Technical Editor Francis Rumsey Staff Technical Writer PURPOSE: The Audio Engineering Society is organized for the purpose of: uniting persons performing professional services in the audio engineering field and its allied arts; collecting, collating, and disseminating scientific knowledge in the field of audio engineering and its allied arts; advancing such science in both theoretical and practical applications; and preparing, publishing, and distributing literature and periodicals relative to the foregoing purposes and policie. MEMBERSHIP: Individuals interested in audio engineering may become members of the AES ( annual dues are: full members and associate members, $149 for both the printed and online Journal; $99 for online Journal only. Student members: $89 for printed and online Journal; $39 for online Journal only. Subscribe to the AES E-Library ( e-lib/subscribe), $145 per year (members), $245 nonmembers. Sustaining memberships are available to persons, corporations, or organizations who wish to support the Society.

2 AES 42 nd INTERNATIONAL CONFERENCE SEMANTIC AUDIO 2011 July COCHAIRS Karlheinz Brandenburg and Mark Sandler PAPERS COCHAIRS Anssi Klapuri and Christian Dittmar TREASURER Sylvia Dietrich SECRETARY Peggy Walther FACILITIES/EXHIBITORS Holger Großmann WEBMASTER Peggy Walther Copyright 2011 Audio Engineering Society, Inc. Library of Congress Catalog Card No ISBN First printing July Printed in the United States of America. For purposes of review or as citation of authority, brief passages may be reproduced from The Proceedings of the AES 42nd International Conference with customary credit to the source. For other purposes, reproduction of any material in The Proceedings of the AES 42nd International Conference requires the permission of the Audio Engineering Society and the author(s). Inquiries should be sent to: Editorial Office, Audio Engineering Society, Inc., 60 East 42nd Street, New York, New York , USA. Telephone: , Fax: The Audio Engineering Society is not responsible for statements made by the contributors. The papers published in these proceedings have been reproduced from the authors manuscripts and without consideration by the Journal Review Board.

3 Contents COMMITTEE S GREETING MUSIC INFORMATION RETRIEVAL PART [Invited Paper] New Developments in Music Information Retrieval...11 Meinard Müller, Saarland University and MPI Informatik, Saarbrücken, Germany 1-2 [Invited Paper] Music Listening in the Future: Augmented Music-Understanding Interfaces and Crowd Music Listening...21 Masataka Goto, National Institute of Advanced Industrial Science and Technology (AIST), Tsukuba, Japan 1-3 [Invited Paper] Lost in Translation: Transferring Semantic Audio Analysis from the Lab to Industry...oral presentation only Jay LeBoef, Imagine Research Inc., San Francisco, CA, USA P POSTER SESSION P1-1 Improving a Multiple Pitch Estimation Method with AR Models...33 Tiago Fernandes Taveres, Jayme Garcia Arnal Barbedo, Amauri Lopes, University of Campinas, Campinas, SP, Brazil P1-2 Polyphonic Music Transcription Using Weighted CQT and Non-Negative Matrix Factorization...39 Sang Ha Park, Seokjin Lee, Koeng-Mo Sung, Seoul National University, Kwanak-gu, Seoul, Korea P1-3 Automatic Classification of Musical Pieces into Global Cultural Areas...44 Anna Kruspe, Hanna Lukashevich, Jakob Abeßer, Holger Großmann, Christian Dittmar, Fraunhofer Institute for Digital Media Technology IDMT, Ilmenau, Germany P1-4 Toward Context-Sensitive Music Recommendations Using Multifaceted User Profiles...54 Rafael Schirru, Andreas Dengel, German Research Center for Artificial Intelligence, Kaiserslautern, Germany and University of Kaiserslautern, Kaiserslautern, Germany; Stephan Baumann, German Research Center for Artificial Intelligence, Kaiserslautern, Germany; Christian Freye, Brandenburg University of Applied Sciences, Brandenburg, Germany P1-5 Regression-Based Tempo Recognition from Chroma and Energy Accents for Slow Audio Recordings...60 Thorsten Deinert, Igor Vatolkin, Günter Rudolph, TU Dortmung, Dortmund, Germany P1-6 Blind Estimation of Reverberation Time from Monophonic Instrument Recordings Based on Non-Negative Matrix Factorization...69 Maximo Cobos, Jose J. Lopez, Universitat Politecnica de Valencia, Valencia, Spain; Pedro Vera- Candeas, Julio Jose Carabias-Orti, Nicolas Ruiz-Reyes, Universidad de Jaen, Linares, Spain AES 42ND INTERNATIONAL CONFERENCE 3

4 Contents continued P1-7 Observing Uncertainty in Music Tagging by Automatic Gaze Tracking...79 Bozena Kostek, Gdansk University of Technology, Gdansk, Poland 2 SPEECH PROCESSING AND ANALYSIS [Invited Paper] Semantic Speech Tagging: Toward Combined Analysis of Speaker Traits...89 Björn Schuller, Martin Wöllmer, Florian Eyben, Gerhard Rigoll, Technische Universität München, Munich, Germany; Dejan Arsic, Müller-BBM Vibroakustiksysteme, Planegg, Germany 2-2 SyncTS: Automatic Synchronization of Speech and Text Documents...98 David Damm, Harald Grohganz, Sebastian Ewert, Michael Clausen, University of Bonn, Bonn, Germany; Frank Kurth, Fraunhofer Institute for Communication, Information Processing and Ergonomics (FKIE), Wachtberg, Germany 2-3 Extraction of Spectro-Temporal Speech Cues for Robust Automatic Speech Recognition Bernd T. Meyer, International Computer Science Institute (ICSI), Berkeley, CA, USA 3 AUTOMATIC MUSIC TRANSCRIPTION Automatic Recognition and Parametrization of Frequency Modulation Techniques in Bass Guitar Recordings Jakob Abeßer, Christian Dittmar, Fraunhofer Institute for Digital Media Technology, IDMT, Ilmenau Germany; Gerald Schuller,Ilmenau University of Technology, Ilmenau, Germany 3-2 Note Clustering Based on 2-D Source-Filter Modeling for Underdetermined Blind Source Separation Martin Spiertz, Volker Gnann, RWTH Aachen University, Aachen, Germany 3-3 Pitch Estimation by the Pair-Wise Evaluation of Spectral Peaks Karin Dressler, Fraunhofer Institute for Digital Media Technology, IDMT, Ilmenau Germany 4 INFORMED SOURCE SEPARATION [Invited Paper] Parametric Coding of Audio Objects: Technology, Performance and Opportunities Jürgen Herre, International Audio Laboratories Erlangen (a joint institution of Fraunhofer IIS and University of Erlangen-Nürnberg), Erlangen, Germany; Leon Terentiv, Fraunhofer Institute for Integrated Circuits IIS, Erlangen, Germany 4-2 Informed Audio Source Separation from Compressed Linear Stereo Mixtures Laurent Girin, Jonathan Pinel, Grenoble Laboratory of Images, Speech, Signal and Automation (GIPSA-lab), Grenoble, France 4-3 Compressive Sensing for Mudic Signals: Comparison of Transforms with Coherent Dictionaries Ch. Srikanth Raj, T. V. Sreenivas, Indian Institute of Science, Bangalor, India 4 AES 42ND INTERNATIONAL CONFERENCE

5 Contents continued 4-4 Sparsification of Audio Signals Using the MDCT/IntMDCT and a Psychoacoustic Model Application to Informed Audio Source Separation Jonathan Pinel, Laurent Girin, Grenoble Laboratory of Images, Speech, Signal and Automation (GIPSA-lab), Grenoble, France P POSTER SESSION P2-1 A Psychoacoustic Approach to Wave Field Synthesis Tim Ziemer, University of Hamburg, Hamburg, Germany P2-2 Comparative Evaluation and Combination of Audio Tempo Estimation Approaches Jose R. Zapata and Emilia Gómez, Universitat Pompeu Fabra, Barcelona, Spain P2-3 Tempo Estimation from Urban Music Using Non-Negative Matrix Factorization Daniel Gärtner, Fraunhofer Institute for Digital Media Technology IDMT, Ilmenau, Germany P2-4 A Musical Source Separation System Using a Source-Filter Model and Beta-Divergence Non-Negative Matrix Factorization Seokjin Lee, SangHa Park, Koeng-Mo Sung, Seoul National University, Kwanak-gu, Seoul, Korea P2-5 Design of a Karaoke System for Commercial Stereophonic Audio Tracks Aiming a Musical Learning Aid for Amateur Singers Karthik. R, Jeyasingh Pathrose, Jasmin Infotech Pvt. Ltd., Chennai, India; M. Madheswaran, Muthayammal Enginering College, Rasipuram, India P2-6 Geometric Source Separation Method of Audio Signals Based on Beamforming and NMFs Seokjin Lee, Sang Ha Park, Koeng-Mo Sung, Seoul National University, Kwanak-gu, Seoul, Korea 5 AUDIO SOURCE SEPARATION Singing Voice Separation from Stereo Recordings Using Spatial Clues and Robust F0 Estimation Pablo Cabañas-Molero, Damián Martínez-Muñoz, University of Jaen, Linares Jaén, Spain; Maximo Cobos, José J. López, Technical University of Valencia, Valencia, Spain 5-2 Influence of Phase, Magnitude and Location of Harmonic Components in the Perceived Quality of Extracted Solo Signals Estefanía Cano, Christian Dittmar, Gerald Schuller, Fraunhofer Institute for Digital Media Technology IDMT, Ilmenau, Germany 6 MUSIC INFORMATION RETRIEVAL PART Expressivity in Musical Timing in Relation to Musical Structure and Interpretation: A Cross-Performance, Audio-Based Approach Cynthia C. S. Liem, Alan Hanjalic, Delft University of Technology, Delft, The Netherlands; Craig Stuart Sapp, CCRMA/CCARH, Stanford University, Stanford, CA, USA AES 42ND INTERNATIONAL CONFERENCE 5

6 Contents continued 6-2 Interactive Classification of Sound Objects for Polyphonic Electro-Acoustic Music Annotation Sebastien Gulluni, Institut National de l Audiovisuel, Bry-sur-marne Cedex, France and Telecom ParisTech, Paris, France; Slim Essid, Gaël Richard, Telecom ParisTech, Paris, France; Olivier Buisson, Institut National de l Audiovisuel, Bry-sur-marne Cedex, France 7 MUSIC INFORMATION RETRIEVAL PART [Invited Paper] Adaptive Distance Measures for Exploration and Structuring of Music Collections Sebastian Stober, Otto-von-Guericke-University Magdeburg, Magdeburg, Germany 7-2 Analyzing Chroma Feature Types for Automated Chord Recognition Nanzhu Jiang, Peter Grosche, Verena Konz, Meinard Müller, Saarland University and MPI Informatik, Saarbrücken, Germany 8 INTELLIGENT AUDIO EFFECTS [Invited Paper] Surround Recording Based on a Coincident Pair of Microphones Christof Faller, ILLUSONIC LLC, St-Sulpice, Switzerland 8-2 A Knowledge Representation Framework for Context-Dependent Audio Processing.305 György Fazekas, Thomas Wilmering, Mark B. Sandler, Queen Mary University of London, London, UK AUTHORS INDEX: AES 42ND INTERNATIONAL CONFERENCE

7 Karlheinz Brandenburg Conference cochair Mark Sandler Conference cochair Committee s Greeting A bout 10 or 12 years ago, it was becoming clear that the next challenge for audio, and more specifically music, to meet full on was the challenge of the Internet. Not just in finding ways to deliver MP3s, which was just starting about then, but in many, many aspects. Since those days, another key technology driver has been the decreasing cost of computing power, coupled with increasing skill in the art and science of digital signal processing. This meant that audio signal processing could get ever more complex, and we scientists and engineers started to talk in terms of features: meaningful, medium-level aspects of signals that captured something abstract about the underlying nature of a signal. We were used to spectral features, such as those from the ubiquitous FFT, but the past 10 years has seen more and more different types of these features introduced, analyzed, criticized, and deployed. And as the complexity of the features has evolved, it has become possible to compute features that have ever stronger correspondence with our understanding (as perceivers and creators) of music and other sources of audio. With understanding comes meaning. Essentially, semantic audio is a 21st century technology. The newest technological addition to semantic representations of content (so not just audio and music but also video, still images, and narrative) is that of the Semantic Web. It enables the concept of ontology to capture specific meaning and relationships between entities in a domain of study and the technology of resource description framework (RDF). These are the most powerful new tools of our ever developing trade. Together they allow the construction of complex online software applications (like mashups but with principles!) and, most important, they allow reasoning to be performed on the representations of the features that the advanced digital signal processing has captured from the original audio source. This means it is now perfectly plausible to search in a large database or collection for songs that have very unique features, such as strong electric guitar in the key of A minor at beats per minute containing no more than four instruments. We are very proud to welcome you to the AES 42nd International Conference on Semantic Audio. The idea for this conference emerged in the AES Technical Committee on Semantic Audio. It would not have been possible without the help of AES headquarters, the AES Conference Policy Committee, and especially everybody on the 42nd Conference Committee. We hope you enjoy fruitful discussions and interactions with everyone attending the conference and have time to sample the cultural and outdoor activities of Ilmenau and the mountains of the Thuringian forest. KARLHEINZ BRANDENBURG AND MARK SANDLER CONFERENCE COCHAIRS AES 42nd INTERNATIONAL CONFERENCE 7