April 19-24, 2009
`
`Proceedings
`
`IEEE Catalog No.: CFP09ICA ISBN: 978-1-4244-2354-5 ISSN: 1520-6149
`
`Page 1 of 7
`
`SONOS EXHIBIT 1029
`
`

`

`2009 IEEE International
`Conference on Acoustics, Speech,
`and Signal Processing
`
`Proceedings
`
`April 19—24, 2009
`Taipei International Convention Center
`Taipei, Taiwan
`
`Sponsored by
`
`The Institute of Electrical and Electronics Engineers
`Signal Processing Society
`
`IEEE Catalog Number: CFP09ICA
`ISBN: 978-1-4244-2354-5
`ISSN: 1520-6149
`
`Page 2 of 7
`
`SONOS EXHIBIT 1029
`
`

`

`Copyright ©2009 by The Institute of Electrical and Electronics Engineers, Inc.
`All rights reserved.
`
`Copyright and Reprint Permission: Abstracting is permitted with credit to the source. Libraries are permitted
`to photocopy beyond the limit of U.S. copyright law for private use of patrons those articles in this volume
`that carry a code at the bottom of the first page, provided the per-copy fee indicated in the code is paid through
`the Copyright Clearance Center, 222 Rosewood Drive, Danvers, MA 01923. For other copying, reprint or
`republication permission, write to IEEE Copyrights Manager, IEEE Operations Center, 445 Hoes Lane, P.O.
`Box 1331, Piscataway, NJ 08855-1331. All rights reserved. Copyright ©2009 by the Institute of Electrical
`and Electronics Engineers, Inc.
`
`The papers in this book comprise the proceedings of the meeting mentioned on the cover and title page. They
`reflect the authors’ opinions and, in the interests of timely dissemination, are published as presented and
`without change. Their inclusion in this publication does not necessarily constitute endorsement by the editors,
`the IEEE Signal Processing Society, or the Institute of Electrical and Electronics Engineers, Inc.
`
`IEEE Catalog Number: CFP09ICA
`ISBN: 978-1-4244-2354-5
`ISSN: 1520-6149
`
`Assembled by Conference Management Services, Inc.
`
`ii
`
`Page 3 of 7
`
`SONOS EXHIBIT 1029
`
`

`

`TABLE OF CONTENTS
`
`AE-L1: AUDIO CODING
`
`AE-L1.1: UNIFIED SPEECH AND AUDIO CODING SCHEME FOR HIGH QUALITY AT LOW .................................... 1
`BITRATES
`Max Neuendorf, Fraunhofer IIS, Germany; Philippe Gournay, University of Sherbrooke, Canada; Markus Multrus, Jérémie
`Lecomte, Fraunhofer IIS, Germany; Bruno Bessette, University of Sherbrooke, Canada; Ralf Geiger, Stefan Bayer, Guillaume
`Fuchs, Johannes Hilpert, Nikolaus Rettelbach, Fraunhofer IIS, Germany; Redwan Salami, VoiceAge Corp., Canada; Gerald
`Schuller, Fraunhofer IDMT, Germany; Roch Lefebvre, University of Sherbrooke, Canada; Bernhard Grill, Fraunhofer IIS,
`Germany
`
`AE-L1.2: AN ERROR ROBUST ULTRA LOW DELAY AUDIO CODER USING AN MA .................................................. 5
`PREDICTION MODEL
`Stefan Wabnik, Fraunhofer, Germany; Gerald Schuller, Ferenc Kraemer, Technical University of Ilmenau, Germany
`
`AE-L1.3: LOW BITRATE AUDIO CODING USING GENERALIZED ADAPTIVE GAIN SHAPE ................................... 9
`VECTOR QUANTIZATION ACROSS CHANNELS
`Sanjeev Mehrotra, Wei-Ge Chen, Kishore Kotteri, Microsoft Corporation, United States
`
`AE-L1.4: AUTOMATIC PARAMETER OPTIMIZATION FOR A PERCEPTUAL AUDIO CODEC ............................... 13
`Martin Holters, Udo Zölzer, Helmut Schmidt University, Germany
`
`AE-L1.5: A MODIFIED DISTORTION METRIC FOR AUDIO CODING ............................................................................. 17
`Vinay Melkote, Kenneth Rose, University of California, Santa Barbara, United States
`
`AE-L1.6: LOW DELAY MOVING-HORIZON MULTIPLE-DESCRIPTION AUDIO CODING ....................................... 21
`FOR WIRELESS HEARING AIDS
`Jan Østergaard, Aalborg University, Denmark; Daniel Quevedo, The University of Newcastle, Australia; Jesper Jensen, Oticon,
`Denmark
`
`AE-L2: SIGNAL ENHANCEMENT AND SOURCE SEPARATION
`
`AE-L2.1: ON NOISE REDUCTION IN THE KARHUNEN-LOEVE EXPANSION DOMAIN............................................. 25
`Jacob Benesty, INRS-EMT, Canada; Jingdong Chen, Bell Labs, Alcatel-Lucent, United States; Yiteng (Arden) Huang, WeVoice,
`Inc., United States
`
`AE-L2.2: MINIMUM SUBSPACE NOISE TRACKING FOR NOISE POWER SPECTRAL .............................................. 29
`DENSITY ESTIMATION
`Mahdi Triki, Kees Janse, Philips Research Laboratories, Netherlands
`
`AE-L2.3: BLIND SPARSE SOURCE SEPARATION FOR UNKNOWN NUMBER OF ...................................................... 33
`SOURCES USING GAUSSIAN MIXTURE MODEL FITTING WITH DIRICHLET PRIOR
`Shoko Araki, Tomohiro Nakatani, Hiroshi Sawada, Shoji Makino, NTT Communication Science Laboratories, Japan
`
`AE-L2.4: BENCHMARKING FLEXIBLE ADAPTIVE TIME-FREQUENCY TRANSFORMS FOR ............................... 37
`UNDERDETERMINED AUDIO SOURCE SEPARATION
`Andrew Nesbit, Queen Mary, University of London, United Kingdom; Emmanuel Vincent, IRISA-INRIA, France; Mark Plumbley,
`Queen Mary, University of London, United Kingdom
`
`AE-L2.5: INTENSITY VECTOR DIRECTION EXPLOITATION FOR EXHAUSTIVE BLIND ....................................... 41
`SOURCE SEPARATION OF CONVOLUTIVE MIXTURES
`Banu Gunel, University of Surrey, United Kingdom; Huseyin Hacihabiboglu, King’s College London, United Kingdom; Ahmet
`Kondoz, University of Surrey, United Kingdom
`
`xix
`
`Page 4 of 7
`
`SONOS EXHIBIT 1029
`
`

`

`IVMSP-P10.3: COMPRESSIVE IMAGING OF COLOR IMAGES ..................................................................................... 1261
`Pradeep Nagesh, Baoxin Li, Arizona State University, United States
`
`IVMSP-P10.4: ROBUST 3D MODELING FROM SILHOUETTE CUES ........................................................................... 1265
`Enliang Zheng, Qiang Chen, Xiaochao Yang, Yuncai Liu, Shanghai Jiao Tong University, China
`
`IVMSP-P10.5: A QUANTITATIVE EVALUATION FOR 3D FACE RECONSTRUCTION ........................................... 1269
`ALGORITHMS
`Vuong Le, Yuxiao Hu, Thomas Huang, University of Illinois at Urbana-Champaign, United States
`
`IVMSP-P10.6: A NOISELESS CODE LENGTH METHOD (NCLM) TO ESTIMATE .................................................... 1273
`DIMENSIONALITY OF HYPERSPECTRAL DATA
`Masoud Farzam, Soosan Beheshti, Ryerson University, Canada
`
`IVMSP-P10.7: NOVEL SIMILARITY INVARIANT FOR SPACE CURVES USING TURNING .................................. 1277
`ANGLES AND ITS APPLICATION TO OBJECT RECOGNITION
`Djamila Aouada, Hamid Krim, North Carolina State University, United States
`
`IVMSP-P10.8: ESTIMATION OF THE HYPERSPECTRAL TUCKER RANKS .............................................................. 1281
`Alexis Huck, Mireille Guillaume, Fresnel Institute, France
`
`IVMSP-P10.9: PANORAMA RECOVERY FROM NOISY UAV SURVEILLANCE VIDEO .......................................... 1285
`Yi Wang, Richard Schultz, Ronald Fevig, University of North Dakota, United States
`
`IVMSP-P10.10: A NEW METHOD TO FIND AN OPTIMAL WARPING FUNCTION IN IMAGE .............................. 1289
`STITCHING
`Hyung Il Koo, Beom Su Kim, Nam Ik Cho, Seoul National University, Republic of Korea
`
`IVMSP-P10.11: INVERSE HALFTONING WITH VARIANCE CLASSIFIED FILTERING .......................................... 1293
`Jing-Ming Guo, Jen-Ho Chen, National Taiwan University of Science and Technology, Taiwan
`
`ITT-L1: SPEECH AND AUDIO PROCESSING APPLICATIONS
`
`ITT-L1.1: EFFICIENT SPEECH INDEXING AND SEARCH FOR EMBEDDED DEVICES ........................................ 1297
`USING UNITERMS
`Changxue Ma, Woojay Jeon, Motorola, United States
`
`ITT-L1.2: A PORTABLE USB-BASED MIROPHONE ARRAY DEVICE FOR ROBUST SPEECH ............................. 1301
`RECOGNITION
`Qi (Peter) Li, Manli Zhu, Li Creative Technologies, Inc., United States; Wei Li, Self-employed, United States
`
`ITT-L1.3: AUDIO-BASED AUTOMATIC MANAGEMENT OF TV COMMERCIALS .................................................. 1305
`Helenca Duxans, David Conejero, Xavier Anguera, Telefónica I+D, Spain
`
`ITT-L1.4: UTTERANCE VERIFICATION USING IMPROVED CONFIDENCE MEASURES .................................... 1309
`BASED ON ALIGNMENT CONFUSION RATE IN CHINESE DIGITS RECOGNITION
`Shilei Zhang, Danning Jiang, Yong Qin, IBM China Research Lab, China
`
`ITT-L1.5: NONLINEAR ACOUSTIC ECHO CONTROL USING AN ACCELEROMETER .......................................... 1313
`Tushar Gupta, Arizona State University, United States; Seth Suppappola, Acoustic Technologies, United States; Andreas Spanias,
`Arizona State University, United States
`
`ITT-L1.6: DOMINANT SPEECH ENHANCEMENT BASED ON SNR-ADAPTIVE SOFT MASK ............................... 1317
`FILTERING
`So-Young Jeong, Jae-Hoon Jeong, Kwang-Cheol Oh, Samsung Electronics Co., Ltd., Republic of Korea
`
`xlii
`
`Page 5 of 7
`
`SONOS EXHIBIT 1029
`
`

`

`SS-L5.5: DECENTRALIZED DYNAMIC SPECTRUM ALLOCATION BASED ON ADAPTIVE ................................ 3645
`ANTENNA ARRAY INTERFERENCE MITIGATION DIVERSITY: ALGORITHMS AND MARKOV
`CHAIN ANALYSIS
`Alexandr Kuzminskiy, Alcatel-Lucent, United Kingdom; Yuri Abramovich, Defence Science and Technology Organization,
`Australia
`
`SS-L5.6: QUICKEST CHANGE DETECTION IN MULTIPLE ON-OFF PROCESSES ................................................... 3649
`Qing Zhao, Jia Ye, University of California, Davis, United States
`
`SS-L6: DISTRIBUTED SIGNAL PROCESSING AND CONSENSUS GOSSIPING
`
`SS-L6.1: DISTRIBUTED SUBGRADIENT PROJECTION ALGORITHM FOR CONVEX ........................................... 3653
`OPTIMIZATION
`Sundhar Ram Srinivasan, Angelia Nedich, Venugopal Veeravalli, University of Illinois at Urbana-Champaign, United States
`
`SS-L6.2: NEIGHBORHOOD GOSSIP: CONCURRENT AVERAGING THROUGH LOCAL ....................................... 3657
`INTERFERENCE
`Bobak Nazer, Alexandros G. Dimakis, Michael Gastpar, University of California, Berkeley, United States
`
`SS-L6.3: INTERVAL CONSENSUS: FROM QUANTIZED GOSSIP TO VOTING .......................................................... 3661
`Florence Benezit, Patrick Thiran, Martin Vetterli, Ecole Polytechnique Federale de Lausanne, Switzerland
`
`SS-L6.4: THE SPEED OF GREED: CHARACTERIZING MYOPIC GOSSIP THROUGH ............................................ 3665
`NETWORK VORACITY
`Deniz Ustebay, Boris Oreshkin, Mark Coates, Michael Rabbat, McGill University, Canada
`
`SS-L6.5: A MIXED TIME-SCALE ALGORITHM FOR DISTRIBUTED PARAMETER ............................................... 3669
`ESTIMATION : NONLINEAR OBSERVATION MODELS AND IMPERFECT COMMUNICATION
`Soummya Kar, José M. F. Moura, Carnegie Mellon University, United States
`
`SS-L6.6: NETWORK GOSSIP ALGORITHMS ...................................................................................................................... 3673
`Devavrat Shah, Massachusetts Institute of Technology, United States
`
`SS-L7: SIGNAL PROCESSING TECHNIQUES AND ALGORITHMS ON ROBOT AUDITION
`
`SS-L7.1: ICA-BASED EFFICIENT BLIND DEREVERBERATION AND ECHO CANCELLATION .......................... 3677
`METHOD FOR BARGE-IN-ABLE ROBOT AUDITION
`Ryu Takeda, Kyoto University, Japan; Kazuhiro Nakadai, Honda Research Institute Japan Co., Ltd., Japan; Toru Takahashi,
`Kazunori Komatani, Tetsuya Ogata, Hiroshi G. Okuno, Kyoto University, Japan
`
`SS-L7.2: SOURCE ADAPTIVE BLIND SIGNAL EXTRACTION USING CLOSED-FORM ICA ................................. 3681
`FOR HANDS-FREE ROBOT SPOKEN DIALOGUE SYSTEM
`Yu Takahashi, Hiroshi Saruwatari, Yuki Fujihara, Kentaro Tachibana, Yoshimitsu Mori, Shigeki Miyabe, Kiyohiro Shikano, Nara
`Institute of Science and Technology, Japan; Akira Tanaka, Hokkaido University, Japan
`
`SS-L7.3: SOUND SOURCE SEPARATION OF MOVING SPEAKERS FOR ROBOT AUDITION................................ 3685
`Kazuhiro Nakadai, Honda Research Institute Japan Co., Ltd./Tokyo Institute of Technology, Japan; Horofumi Nakajima, Yuji
`Hasegawa, Hiroshi Tsujino, Honda Research Institute Japan Co., Ltd., Japan
`
`SS-L7.4: 2D SOUND SOURCE MAPPING FROM MOBILE ROBOT USING ................................................................. 3689
`BEAMFORMING AND PARTICLE FILTERING
`Satoshi Kagami, Simon Thompson, National Institute of Advanced Industrial Science and Technology (AIST), Japan; Yoko
`Sasaki, Hiroshi Mizoguchi, Tokyo University of Science, Japan; Tadashi Enomoto, Kansai Electric Power Co., Inc., Japan
`
`SS-L7.5: DOA ESTIMATION METHOD BASED ON SPARSENESS OF SPEECH SOURCES ..................................... 3693
`FOR HUMAN SYMBIOTIC ROBOTS
`Masahito Togami, Akio Amano, Takashi Sumiyoshi, Yasunari Obuchi, Hitachi Ltd., Japan
`
`lxxxvi
`
`Page 6 of 7
`
`SONOS EXHIBIT 1029
`
`

`

`SS-L7.6: A SINGLE-CHIP SPEECH DIALOGUE MODULE AND ITS EVALUATION ON A ...................................... 3697
`PERSONAL ROBOT, PAPERO-MINI
`Miki Sato, Toru Iwasawa, Akihiko Sugiyama, Toshihiro Nishizawa, Yosuke Takano, NEC Corporation, Japan
`
`SS-L8: THE DATA DELUGE: THE CHALLENGES AND OPPORTUNITIES OF UNLIMITED
`DATA IN SIGNAL PROCESSING
`
`SS-L8.1: THE DATA DELUGE: CHALLENGES AND OPPORTUNITIES OF UNLIMITED ....................................... 3701
`DATA IN STATISTICAL SIGNAL PROCESSING
`Michael L. Seltzer, Microsoft Research, United States; Lei Zhang, Microsoft Research Asia, China
`
`SS-L8.2: FILTERING WEB TEXT TO MATCH TARGET GENRES ................................................................................. 3705
`Alex Marin, Sergey Feldman, Mari Ostendorf, Maya Gupta, University of Washington, United States
`
`SS-L8.3: LARGE SCALE NATURAL IMAGE CLASSIFICATION BY SPARSITY EXPLORATION .......................... 3709
`Changhu Wang, University of Science and Technology of China, China; Shuicheng Yan, National University of Singapore,
`Singapore; Hong-Jiang Zhang, Microsoft Advanced Technology Center, China
`
`SS-L8.4: LEVERAGING MULTIPLE QUERY LOGS TO IMPROVE LANGUAGE MODELS .................................... 3713
`FOR SPOKEN QUERY RECOGNITION
`Xiao Li, Patrick Nguyen, Geoffrey Zweig, Dan Bohus, Microsoft Research, United States
`
`SS-L8.5: ANNOTATING IMAGES BY HARNESSING WORLDWIDE USER-TAGGED .............................................. 3717
`PHOTOS
`Xirong Li, Cees Snoek, Marcel Worring, University of Amsterdam, Netherlands
`
`SS-L8.6: CO-ADAPTATION: ADAPTIVE CO-TRAINING FOR SEMI-SUPERVISED LEARNING ............................ 3721
`Gokhan Tur, SRI International, United States
`
`SS-L9: HANDLING REVERBERANT SPEECH: METHODOLOGIES AND APPLICATIONS
`
`SS-L9.1: STRATEGIES FOR MODELING REVERBERANT SPEECH IN THE FEATURE ........................................ 3725
`DOMAIN
`Armin Sehr, Walter Kellermann, University of Erlangen-Nuremberg, Germany
`
`SS-L9.2: HANDS-FREE SPEECH RECOGNITION CHALLENGE FOR REAL-WORLD ............................................ 3729
`SPEECH DIALOGUE SYSTEMS
`Hiroshi Saruwatari, Hiromichi Kawanami, Shota Takeuchi, Yu Takahashi, Tobias Cincarek, Kiyohiro Shikano, Nara Institute of
`Science and Technology, Japan
`
`SS-L9.3: ADAPTIVE DEREVERBERATION OF SPEECH SIGNALS WITH .................................................................. 3733
`SPEAKER-POSITION CHANGE DETECTION
`Takuya Yoshioka, Hideyuki Tachibana, Tomohiro Nakatani, Masato Miyoshi, Nippon Telegraph and Telephone Corporation,
`Japan
`
`SS-L9.4: BLIND SYSTEM IDENTIFICATION FOR SPEECH DEREVERBERATION WITH ..................................... 3737
`FORCED SPECTRAL DIVERSITY
`Xiang (Shawn) Lin, Imperial College London, United Kingdom; Andy W. H. Khong, Nanyang Technological University,
`Singapore; Patrick A. Naylor, Imperial College London, United Kingdom
`
`SS-L9.5: ON A TRADEOFF BETWEEN DEREVERBERATION AND NOISE REDUCTION ...................................... 3741
`USING THE MVDR BEAMFORMER
`Emanuël Habets, Technion - Israel Institute of Technology, Israel; Jacob Benesty, University of Quebec, Canada; Israel Cohen,
`Technion - Israel Institute of Technology, Israel; Sharon Gannot, Bar-Ilan University, Israel
`
`SS-L9.6: ROOM IMPULSE RESPONSE SHORTENING WITH INFINITY-NORM ....................................................... 3745
`OPTIMIZATION
`Tiemin Mei, Alfred Mertins, Markus Kallinger, University of Luebeck, Germany
`
`lxxxvii
`
`Page 7 of 7
`
`SONOS EXHIBIT 1029
`
`