UNITED STATES PATENT AND TRADEMARK OFFICE
`
`_________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`_________________
`
`GOOGLE LLC,
`Petitioner,
`
`v.
`
`BUFFALO PATENTS, LLC,
`Patent Owner.
`
`_________________
`
`Case No. IPR2023-01387
`U.S. Patent No. 8,204,737
`_________________
`
`Declaration of Shauna L. Wiest Regarding Bellegarda
`
`Page 1 of 72
`
`GOOGLE EXHIBIT 1019
`
`

`

`
`
`Declaration of Shauna L. Wiest
`
`
`I, Shauna L. Wiest, state and declare as follows:
`
`I.
`
`Introduction
`
`1.
`
`I have prepared this Declaration in connection with Google LLC’s
`
`(“Petitioner”) Petition for Inter Partes Review of U.S. Patent No. 8,204,737,
`
`which I understand will be filed concurrently with this Declaration.
`
`2.
`
`I am a senior research analyst with the Research & Information
`
`Services team at Finnegan, Henderson, Farabow, Garrett & Dunner, LLP located at
`
`3300 Hillview Avenue, Palo Alto, CA 94304 (“Finnegan”).
`
`3.
`
`I am over eighteen years of age, and I am competent to make this
`
`Declaration. I make this Declaration based on my own personal knowledge,
`
`and my professional knowledge of library science practices.
`
`4.
`
`I earned a Master of Science in Library Science degree from the
`
`University of North Carolina at Chapel Hill in 1999, and a Bachelor of Arts in
`
`Political Science degree from the University of California at San Diego in
`
`1989. I have worked as a law librarian for over eighteen years. I have been
`
`employed in the Research & Information Services Department at Finnegan
`
`since 2021. Before that, from 2000-2015, I was employed as a Law Librarian
`
`at Stoel Rives LLP, and from 2015-2016, I was employed as a Competitive
`
`Intelligence Specialist for Nossaman LLP.
`
`
`
`2
`
`Page 2 of 72
`
`

`

`
`II.
`
`Declaration of Shauna L. Wiest
`
`
`Standard Library Practice for Receiving, Cataloging, Shelving, and
`Making Materials, including Conference Publications, Publicly
`Available
`
`5.
`
`I have knowledge of and experience with standard library practices
`
`regarding receiving, cataloging, shelving, and making materials, including
`
`conference publications, available to the public. I am fully familiar with and
`
`have knowledge of and experience with the Machine-Readable Cataloging
`
`(MARC) system, an industry-wide standard that libraries use to catalog
`
`materials.
`
`6.
`
`The MARC system was developed during the 1960s to standardize
`
`bibliographic catalog records so they could be read by computers and shared
`
`among libraries. By the mid-1970s, MARC had become the international standard
`
`for cataloging bibliographic materials and is still used today. Many libraries
`
`provide public access to their MARC records via the Internet and/or their
`
`electronic cataloging systems at the library. In a MARC record, each field provides
`
`specific information about the cataloged item, including how materials are held
`
`and made available to the public.
`
`III. MARC Records
`
`7.
`
`The MARC record system uses a specific three-digit numeric code
`
`(“field tags”) (from 001-999) to identify each field in a catalog record. For
`
`example, MARC field tag 008 provides the six-digit date the item was received
`
`
`
`3
`
`Page 3 of 72
`
`

`

`Declaration of Shauna L. Wiest
`
`and catalogued (Date entered on file). The first six characters of field tag 008 are
`
`always in the “YYMMDD” format. Descriptions and definitions of all of the
`
`character positions of field tag 008 are outlined here:
`
`https://www.loc.gov/marc/bibliographic/bd008a.html
`
`8.
`
`As is relevant to this Declaration, MARC field tag 245 identifies the
`
`title and statement of responsibility for the work. MARC field tag 260 identifies
`
`the place of publication, name of publisher, and date of the publication. MARC
`
`field tag 310 sets forth the current publication frequency for a work. MARC field
`
`tag 362 sets forth the dates of publication and/or sequential date(s) of publication
`
`designation for a work. MARC field tag 515 provides any numbering or publishing
`
`pattern peculiarities for the resource. MARC field tag 991 provides local holdings
`
`information to assist the public in identifying the location of a desired resource.
`
`9.
`
`Based on standard library practice, when a library receives an item,
`
`it generally stamps (and/or labels) the item with the library name, barcode,
`
`often with a date that is within a few days or weeks of receipt. Next, the library
`
`will catalog the item within a matter of a few days or weeks of receiving it. As
`
`a general practice, cataloguing is centralized and performed by a cataloguing
`
`department within a library or university setting. In certain circumstances the
`
`catalogued item may be subsequently sent to a library location within the
`
`4
`
`Page 4 of 72
`
`

`

`Declaration of Shauna L. Wiest
`
`library or university setting where it may be stamped and/or labeled after it has
`
`been catalogued.
`
`10. Generally, after an item is cataloged, the public may access the
`
`item by searching a library catalog, browsing the library shelves, and either
`
`requesting or electronically accessing the item from the library. Standard
`
`library practice is to make the item available to the public within a few days or
`
`weeks of cataloging it.
`
`IV. Cataloguing of Conference Proceedings (Monograph v. Serial
`Treatment)
`
`11. Conference proceedings often consist of collections of papers
`
`presented at a meeting, program, symposium or conference held during a
`
`specific time-period in a designated geographic place. Bibliographic control of
`
`conference proceedings to promote public discovery and access has long been
`
`a priority of catalogers. Although some ongoing conference publications are
`
`catalogued as serials (if a conference publication exhibits continuity,
`
`regularity, and evidence of seriality), cataloguing a conference publication as a
`
`monograph is preferred for public retrieval purposes. Cataloguing a conference
`
`publication as a monograph is an optimal way for the public to easily access
`
`the conference resource, and its papers, by searching for a unique title and
`
`other descriptive data elements added by the cataloger. Monographs are
`
`efficient for public searching and offer the public an easier way to discover
`
`5
`
`Page 5 of 72
`
`

`

`Declaration of Shauna L. Wiest
`
`what an institution holds, while changes to the record (unlike serials) are not an
`
`issue. However, sometimes conference proceeding cataloging is done serially,
`
`and the MARC record will be updated after each serial publication is received.
`
`V.
`
`Public Availability of Bellegarda
`
`12.
`
`This Declaration relates to the dates of receipt and public
`
`availability of the following reference: J. R. Bellegarda, J. W. Butzberger,
`
`Yen-Lu Chow, N. B. Coccaro and D. Naik, “A novel word clustering algorithm
`
`based on latent semantic analysis,” (“Bellegarda”) in the 1996 IEEE
`
`International Conference on Acoustics, Speech, and Signal Processing
`
`Conference Proceedings, May 7-10, 1996, Atlanta, GA, USA, pp. 172-175
`
`volume 1 (“1996 IEEE Publication”). I understand that Bellegarda has been
`
`submitted as Exhibit 1009 in this proceeding. That same reference is appended
`
`to my Declaration as Appendix A.
`
`13. As detailed below, I have reviewed the print reference, public
`
`holdings information, and Library of Congress MARC record for Bellegarda to
`
`determine the date of public availability for this reference.
`
`14. Appendix A to this Declaration is a true and accurate copy of the
`
`cover page (including binding), title page, front matter, table of contents, call
`
`number, and date stamp for the copy of the 1996 IEEE Publication containing
`
`Bellegarda held by the Library of Congress. The date stamp on the copy of 1996
`
`6
`
`Page 6 of 72
`
`

`

`Declaration of Shauna L. Wiest
`
`
`IEEE Publication containing Bellegarda indicates that Bellegarda was received by
`
`the Library of Congress on September 11, 1996. The binding indicates that the
`
`associated Call Number is TK7882.S65 I37a. Appendix A also includes pages
`
`172-175, which is the specific article entitled “A novel word clustering algorithm
`
`based on latent semantic analysis),” i.e., Bellegarda.
`
`15. Appendix B to this Declaration is a true and accurate copy of the
`
`Library of Congress public catalog record for its copy of 1996 IEEE Publication
`
`containing Bellegarda, which was downloaded from https://lccn.loc.gov/98655050
`
`on August 30, 2023. The Library of Congress public catalog record sets forth the
`
`public holdings and onsite location information for members of the public seeking
`
`a physical copy of the 1996 IEEE Publication containing Bellegarda. The public
`
`catalog record indicates that the copy of the 1996 IEEE Publication containing
`
`Bellegarda should be requested in the Jefferson or Adamas Building Reading
`
`Rooms at Call Number TK7882.S65 I37a. The public catalog record notes that the
`
`publications associated with the record are (i) published “annual[ly],” (ii)
`
`published in 1995 and 1996, (iii) issued in five (5) or more parts, and (iv) should
`
`be requested in the Jefferson or Adams Building Reading Rooms at Call Number
`
`TK7882.S65 I37a. This information indicates serial cataloguing, as discussed
`
`above in paragraph 11. The public catalog record also notes a publication location
`
`in Piscataway, NJ, with copyright dates of 1995 and 1996, and IEEE as the
`
`
`
`7
`
`Page 7 of 72
`
`

`

`Declaration of Shauna L. Wiest
`
`publisher. Based on my experience as a librarian, the public catalog record
`
`(Appendix B) references the 1996 IEEE Publication containing Bellegarda
`
`(Appendix A).
`
`16. Appendix C to this Declaration is a true and accurate copy of the
`
`Library of Congress MARC record for its holdings of 1996 IEEE Publication
`
`containing Bellegarda, which was downloaded from
`
`https://catalog.loc.gov/vwebv/staffView?searchId=23158&recPointer=0&recCount
`
`=25&searchType=1&bibId=11501706 on August 30, 2023. The Library of
`
`Congress MARC record field tag 245 identifies the full title statement for the work
`
`as: |a Conference proceedings / |c the ... International Conference on Acoustics,
`
`Speech, and Signal Processing; sponsored by the Signal Processing Society of the
`
`Institute of Electrical and Electronics Engineers. MARC field tag 260 sets forth the
`
`place of publication, publisher, and publication dates as: |a Piscataway, NJ : |b
`
`Institute of Electrical and Electronics Engineers, |c c1995-c1996. MARC field tag
`
`310 indicates that the IEEE International Conference on Acoustics, Speech, and
`
`Signal Processing Conference Proceedings is an annual publication. MARC field
`
`tag 362 indicates the sequential dates of publication for the resource as 1995-1996,
`
`indicating publication dates of both 1995 and 1996. MARC field tag 515 denotes
`
`numbering peculiarities for the resource as being issued in 5 or more parts. Finally,
`
`MARC field tags 991 set forth the Library of Congress local holdings and call
`
`8
`
`Page 8 of 72
`
`

`

`Declaration of Shauna L. Wiest
`
`number information for Bellegarda as General Collection, Call Number
`
`TK7882.S65 I37a (Serials). Based on my experience as a librarian, the MARC
`
`record (Appendix C) references the 1996 IEEE Publication containing Bellegarda
`
`(Appendix A).
`
`17. Appendix C confirms the fixed data elements of MARC field tag 008
`
`as 950630d19951996njuar 1 0eng c. Data elements “d1995-1996” in field tag 008
`
`indicate (i) issues of the publication associated with this MARC record were
`
`received during 1995 and 1996, and (ii) the first volume of the publication
`
`associated with the MARC record is dated 1995, and the last publication is dated
`
`1996. As discussed above in paragraph 7, the first six characters “950630” are in
`
`typical “YYMMDD” format and indicate that the first volume of publication for
`
`the resource associated with the MARC record was received by The Library of
`
`Congress on June 30, 1995.
`
`18. Based on the information in Appendices A, B, and C, the 1996 IEEE
`
`Publication containing Bellegarda was received by the Library of Congress on
`
`September 11, 1996. Based on standard library practices, Bellegarda would have
`
`been processed and catalogued by the Library of Congress within a matter of a few
`
`days or weeks of September 11, 1996.
`
`19. Accordingly, Bellegarda would have been made available to the
`
`public within a few days or weeks of being checked-in and catalogued. The public
`
`9
`
`Page 9 of 72
`
`

`

`Declaration of Shauna L. Wiest
`
`could have accessed Bellegarda within a few weeks of September 11, 1996 by (i)
`
`searching the Library of Congress online catalog by, for example, searching by
`
`title, author, and/or subject keywords, or (ii) by asking a library staff member and
`
`being directed to the Jefferson or Adamas Building Reading Rooms and/or General
`
`Collections at Call Number TK7882.S65 I37a.
`
`VI. Conclusion
`
`20.
`
`In signing this Declaration, I understand it will be filed as evidence
`
`in a contested case before the Patent Trial and Appeal Board of the United
`
`States Patent and Trademark Office. I understand I may be subject to cross-
`
`examination in this case and that cross-examination will take place within the
`
`United States. If cross-examination is required of me, I will appear for cross-
`
`examination within the United States during the time allotted for cross-
`
`examination.
`
`21.
`
`I declare that all statements made herein of my knowledge are true,
`
`that all statements made on information and belief are believed to be true, and that
`
`these statements were made with the knowledge that willful false statements and
`
`the like so made are punishable by fine or imprisonment, or both, under Section
`
`1001 of Title 18 of the United States Code.
`
`10
`
`Page 10 of 72
`
`

`

`Executed on September 11, 2023.
`
`Declaration of Shauna L. Wiest
`
`Shauna L. Wiest
`
`11
`
`Page 11 of 72
`
`

`

`APPENDIX A
`APPENDIX A
`
`Page 12 of 72
`
`Page 12 of 72
`
`

`

`SSVOI
`
`Page 13 of 72
`
`BEEAIOE
`
`

`

`Conference Proceedings
`
`May7 - 10, 1996
`Atlanta, Georgia USA
`
`1996 IEEE International Conference on Acoustics,
`Speech, &Signal Processing
`
`Institute of Electrical and Electronic Engineers
`
`AVYTTTTIom |
`
`Sponsored bythe
`Signal Processing Society of the
`
`Page 14 of 72
`
`Page 14 of 72
`
`

`

`LCASS#)
`
`
`The 1996 IEEE International Conference on
`Acoustics, Speech, and Signal Processing
`Conference Proceedings
`_
`
`»
`
`Sponsored by the Signal Processing Society of the Institute of Electrical and
`Electronics Engineers
`
`May 7-10, 1996
`Marriott Marquis Hotel
`Atlanta, Georgia, USA
`
`Page 15 of 72
`
`Page 15 of 72
`
`

`

`The 1996 IEEE International Conference on Acoustics, Speech, and Signal Processing
`Conference Proceedings
`
`Copyright and Reprint Permission: Abstracting is permitted with credit to the source. Libraries are permitted to photocopy
`beyondthe limit of U.S. copyright law for private use of patrons those articles in this volumethat carry a codeat the bottom
`ofthe first page, provided the per-copy fee indicated in the codeis paid through Copyright Clearance Center, 222 Rosewood
`Drive, Danvers, MA 01923. For other copying, reprint or republication permission, write to IEEEa
`Ci =
`[KIEE HY
`_
`YO
`65_T3%,
`V DIGG bE
`
`Copyrights Manager, IEEE Service Center, 445 Hoes Lane, P.O. Box 1331, Piscataway, NJ
`08855-1331. All rights reserved. Copyright 1996by the Institute of Electrical and Electronics
`Engineers, Inc.
`
`96CH35903
`IEEE Catalog Number:
`ISBN 0-7803-3192-3 (softhound)
`ISBN 0-7806-3193-1 (casebound edition)
`ISBN 0-7803-3194-X (microfiche)
`ISBN 0-7803-3195-8 (CD-ROM)
`Library of Congress:
`84-645 139
`
`Additional Proceedings (hard-copy and CD-ROM)maybe ordered from:
`
`IEEE Service Center
`445 Hocs Lane
`P.O, Box 1331
`Piscataway, NJ 08855-1331
`1-800-678-IEEE
`
`Page 16 of 72
`
`i
`
`Page 16 of 72
`
`

`

`Volume 1
`
`TABLE OF CONTENTS
`
`SP1 Robust Recognition: Signals and Features
`Feature Parameter Curve Method for High Performance NN-based Speech Recognition .........scssssseersssreencseeeserenenensnsensns I-1
`D. Chen, S. Zhu, T: Huang - Chinese Academy of Sciences, China
`
`Volume I
`
`On the Use of Residual Cepstrum im Speech Recognition—.....cccsssceesscccsusesereeusereuanenseeeseueeeeeussesenseenensees sisdebesebenesexeass1-5
`J. He, L. Liu, G. Palm - University of Ulm, Germany
`
`RobustDistant Talking Speech Recognition—....sssecccessseseeesnecsenees SetueddsevaaaacerNeaaTeeae's Serer ere ress sescvccceseveseeeol-21
`Q. Lin, C. Che, D. Yuk, L. Jin - CAIP Center, Rutgers University, USA
`B. Vries, J. Pearson - David SarnoffResearch Center, USA
`J. Flanagan - Rutgers University, USA
`
`HMM.-BasedSpeech Recognition Using State-Dependent, Linear Transforms on Mel-Warped DFT Features
`C. Rathinavelu, L. Deng - University of Waterloo, Canada
`
` ...ssssccecessevseees I-9
`
`Mixed Malvar-Wavelets for Non-Stationary Signal Representation .......00sssse0e+ ibid pahdoases ddsndeetuscndaennsepasenthanasspicasssatessl-13
`J. Thripuraneni, W. Lou, V. DeBrunner - The University of Oklahoma, USA
`
`Experiments on a Parametric Nonlinear Spectral Warping for an HMM-based Speech Recognizer ...........ssscccesesccsesoveeeee I-L7
`D. Mashao - Brown University, USA
`
`Time-Frequency Representation Based Cepstral Processing For Speech Recognition .........ssssscccecesseseeesseceessevensanseeeasee I-25
`A. Fineberg, K. Yu - Motorola Lexicus, USA
`
`Knowledge-Based Parameters for Speech HMM Recognition ...........ccececsersentesee seelanes STP TTTT saaaaanesenaewesaseaieas seeeeseeed-29
`C. Espy- Wilson, N. Bitar - Boston University, USA
`oe
`
`A PhonemeSimilarity Based ASR Front-End .....sccsscesecsseeensneceeeanecerenseeeseneurssenseeeteuesnseneuaneneneceueseteeeecesssseeesseesees «I-33
`T. Applebaum, P. Morin, B. Hanson - Speech Technology Laboratory, USA
`
`A Model of Dynamic Auditory Perception andits Application to Robust Speech Recognition ....4......0..sesesssseseesreenssenseeneI-37
`B. Strope, A. Alwan - University of California at Los Angeles, USA
`
`SP2 Robust Recognition: Large Vocabulary
`Independent Calculation of Power Parameters on PMC Method............. debaeebaue haeelendbancuersdetebencis Wa tiaTeTbias te tebaeaTETeRats:[-41
`Hf. Yamamoto, M. Yamada, T. Kosaka, ¥. Komori, ¥. Ohora - Canon Inc., Japan
`
`Noisy Speech Recognition Using Variance Adapted Likelihood Measure—...csssesssecssssveneccseenscseesensesssnsseserseseeneesssenseses “G5
`J. Chien, L. Lee, H. Wang - National Tsing Hua University, ROC
`
`An Improved Noise Compensation Algorithm for Speech Recognition in Noise .....sscsssseecsseserenerersesseveenensensensserenssereneesd-AO
`R. Yang, P. Haavisto - Nokia Research Center, Finland
`
`Improved Speech Recognition via Speaker Stress Directed Classification ..cccccescsesesecsseensecnecsveceeensevenensresasseeeeeesecnaeeee1-53
`B. Womack, J. Hansen - Duke University, USA
`
`High-Accuracy Connected Digit Recognition for Mobile Applications
`S. Gupta, F Soong, R. Haimi-Cohen - AT&T Bell Labs, USA
`
`..ccccccesesecenseecuseeuceeuseseeeeeersuseseneeeeeeaeeneneaseeeaazerI-57
`
`Feature Extraction Based on Zero-Crossings with Peak Amplitudes for Robust Speech Recognition in Noisy Environments... I-61
`D. Kim, J. Jeong, J. Kim, S. Lee - Korea Advanced Institute of Science and Technology, ROK
`
`Improving Environmental Robustness in Large Vocabulary Speech Recognition .......:+s000+ sePeNvereusuevereeneneevausesereesserenens 1-65
`P. Woodland, M. Gales, D, Pye - Cambridge University, UK
`
`Noise and Room Acoustics Distorted Speech Recognition by HMM Composition ....sisscssssesseensecesssssereeeesssvessseneeesserenneedQ9
`S. Nakamura, T. Takiguchi, K. Shikano - Nara Institute of Science and Technology, Japan
`
`Developments in Continuous Speech Dictation Using the 1995 ARPA NAB News Task ....cssccssccssreeeesseeeseneneenseeeeereesseres «I-73
`J. Gauvain, L. Lamel, G. Adda, D. Matrouf - LIMSI-CNRS, France
`
`Evaluation of the Root-Normalized Front-End (RN-LFCC) for Speech Recognition in Wireless GSM Network Environments I-77
`P. Lockwood, S. Dufour, C. Glorion - MATRA Communications, France
`
`Page 17 of 72
`
`Page 17 of 72
`
`

`

`VolumeI
`[aecacoy
`
`SP3 Speaker Recognition
`Speaker Background Models for Connected Digit Password Speaker Verification—.ssssseseesssrasenssernensrerereess eaeeeeee nesI-81
`A. Rosenberg, S. Parthasarathy - AT&T Bell Laboratories, USA
`
`CohortSelection and Word GrammarEffects on Speaker Recognition .....ssseecssereeeeereerenens sendgugevnadenssaeanctennan seeneeneeree el=B85
`J. Colombi, D. Ruck - AFIT/ENG, USA,
`T. Anderson - AL/CFBA, USA,
`S. Rogers - AFIT/ENG, USA,
`M. Oxley - AFIT/ENC, USA
`
`Discriminative Training of GMM for Speaker Identification .........csssesccnseeteoneesescausonscasccesssguuaqavensnssnvgsunsencsssdiiii .-. 1-89
`C. Martin Del Alamo, J. Caminero Gil, C. De La Torre, L. Hernandez Gomez - Telefonica I+D, Spain
`Subword-based Text-dependent Speaker Verification System with User-Selectable Passwords—..sssssssssseseseserersennnrnnanens veee1-93
`M. Sharma, R. Mammone - Rutgers University, USA
`
`Robust Methods of Updating Model and A Priori Threshold in Speaker Verification
`T. Matsui, T. Nishitani, S. Furui - NTT Human Interface Laboratories, Japan
`
`...sessecssesesssessaseseneenes Sewbeaensaaaewaveene1-97
`
`A FurtherInvestigation of AR-Vector Models for Text-Independent Speaker Identification ..........secccccsreseeserrenneenrenaans I-101
`I. Magrin-Chagnolleau, J. Wilke, F. Bimbot - CNRS, France
`
`Speaker Identification via Support Vector Classifiers ........sccscsssnsessesseeeetenes sesbesssaseeeresaeys ceaseeeesenceceetsenerenenseeeeeeese n=LOS
`M. Schmidt - BBN, USA
`
`Speaker Verification Using Mixture Likelihood Profiles Extracted from Speaker Independent Hidden Markov Models
`A. Setlur, R. Sukkar, M. Gandhi - AT&T Beil Laboratories, USA
`
`......1-109
`
`The Effects of Handset Variability on Speaker Recognition Performance: Experiments on the Switchboard Corpus ......... 1-113
`D. Reynolds - MIT Lincoln Laboratory, USA
`
` ......:ss0csessseesenesneees sasbesesnensceveusssusensss neseuaenageeasyyeiias sussecesecesececssoeh@117
`Speaker Recognition in Reverberant Enclosures
`P. Castellano, §. Sridharan, D. Cole - Signal Processing Research Centre, Australia
`
`Speech Recognition: Noise and Environment
`SP4
`Using a Transcription Graph for Large Vocabulary Continuous Speech Recognition ..........ssseeeeereeeneresas eeaseeeeeneesceneeees -I-121
`Z. Li, D, O'Shaughnessy - INRS-Telecommunications, Canada
`Fast and Accurate Recognition ofVery-Large-Vocabulary Continuous MandarinSpeech for Chinese Language with Improved
`Segmental Probability Modeling ........ sibaas vias bie toueks! sn WAM eR WA MeNeaNATENDAESONaReNEMAEREET EABOcoee seceescrecescsevesscsecsccsessssecesevalLZ5
`J. Shen, S. Hwang - National Taiwan University, ROC
`L, Lin-shan - Academia Sinica, ROC
`
`Decoding Optimal State Sequence with Smooth State Likelihoods.......... es cncavecesssescaseusagsausasarecarauscererssessensntanee anaaeven 1-129
`I. Zeljkovic - AT&T Bell Laboratories, USA
`
`Improvements on the Pronunciation Prefix Tree Search Organization .......seccsccssesssssssvstsesveressseeassssseoonnssseversrrereseenes1-133
`FE. Alleva, X. Huang, M. Hwang - Microsoft Corporation, USA
`
`Minimizing Search Errors Due to Delayed Bigrams in Real-Time Speech Recognition Systems
`M. Woszczyna - University of Karlruhe, Germany, M. Finke - University of Karlsruhe, Germany
`
`......:ssscseessesssessrereess seveee 1-137
`
`Real-Time Recognition of Broadcast Radio Speech .........064 peeecaseneapecscccaesseconecesenreoonersroncssssessesceesauan songs Fiveveapyievel1-141
`G. Cook, J. Christie, P. Clarkson, S. Cooper, M. Hochberg, D. Kershaw, R. Logan, S. Renals, A. Robinson,
`C. Seymour, S. Waterhouse, P. Zolfaghari - Cambridge University, UK
`Spontaneous Dialogue Speech Recognition Using Cross-Word Context Constrained Word Graphs..... secencaavanebbeavaeba sevesed-145
`T. Shimizu, H. Yamamoto, H. Masataki, S. Matsunaga, Y. Sagisaka - ATR - ITL, Japan
`
`......... oa sadeunietavies giceetcaaseabeeecheuvassseveeee [-149
`Efficient Evaluation of the LYVCSR Search Space Using the NOWAYDecoder
`S. Renals - University of Sheffield, UK, M. Hochberg - University of Cambridge, UK
`Developments in Large Vocabulary, Continuous Speech Recognition of German ......... aqanananpeenagnasiendiandbepanbe Sdaaacessnes . 1-153
`M. Adda-Decker, G. Adda, L. Lamel, J. Gauvain - LIMSI-CNRS, France
`
`Speech Recognition on Mandarin Call Home: A Large-Vocabulary, Conversational, and Telephone Speech Corpus_....++..1-157
`ELiu, M. Picheny, P. Srinivasa, M. Monkowski, J. Chen - IBM T.J. Watson Research Center, USA
`
`xii
`
`Page 18 of 72
`
`Page 18 of 72
`
`

`

`Volume I
`2Seolaea.
`
`Multilingual Stochastic n-Gram Class Language Models_......:.ssccsecccorsootonsnseausenansusvencceassasesseaceeesssseeeaaeeeeeegsaeenauee 1-161
`M. Jardino - LIMSI-CNRS, France
`
`SP5 Speech-Recognition: Language Modeling IIl
`
`A Variable-Length Category-Based N-Gram Language Model
`T. Niesler, P. Woodland - Cambridge University, UK
`
` ......ssscsssoscessssevecsctersescerensectsncscusteccecseseuccecescucersuanes 1-164
`
`Improving N-Gram Models by Incorporating Enhanced Distributions ............000.. orn sbesseceseceseserccencreccacsensoccess bbdaued 1-168
`P. OBoyle, J. Ming, J. McMahon, J. Smith - Queen's University of Belfast, UK
`
`A Novel Word Clustering Algorithm Based on Latent Semantic Analysis ......:sssecsesssencssccnensceteseenseenenesceenensctecescuanaaseeT-172
`J. Bellegarda, J. Butzberger, Y. Chow, N. Coccaro, D. Naik - Interactive Media Group, Apple Computer USA
`
`Statistical Natural Language Understanding.......s0ssscssseresssessecerenee Aseeeeeeeeaseneeransenunrsesenaes
`sevessscenecssceeeeasl=176
`M. Epstein, K. Papineri, S. Roukos, T. Ward - IBM TJ. Watson Research Center, USA,S.‘Della Pietra - Renaissance:Technologies,"USA
`Clustering Wordsfor Statistical Language Models Based on Contextual Word Similarity .......s.cccesssseseeseseneesee 1-180
`A. Farhat, J. Isabelle, D. O'Shaughnessy - INRS-Telecommunications, Canada
`
`Domain Word Translation By Space-Frequency Analysis of Context Length Histograms ............ seseeeneeeereneeeesecesarseseeeeel-LO4
`P. Fung - Columbia University, USA
`
`Variable-Order N-Gram Generation by Word-class Splitting and Consecutive Word Sequence Grouping ........sseseeseseres oes 1-188
`H. Masataki, Y. Sagisaka - ATR Interpreting Telecommunications Research Laboratories, Japan
`
`Back-off Method for N-Gram Smoothing Based on Binomial Posteriori Distribution ...........0ccssscostesvessscscanscens sacocccsvceee 1-192
`T. Kawabata, M. Tamoto - NTT Basic Research Labs, Japan
`
`Ergodic Multigram HMMIntegrating Word Segmentation and Class Tagging for Chinese Language Modeting oovdébersueriie 1-196
`
`H. Law, C. Chan - The University of Hong Kong, Hong Kong
`
`SP6 Low-Rate Speech Coding
`A 2.4 kbit/s MELP Coder Candidate for the New U.S. Federal Standard .........:.cccsccccseeseseeesscyecsessscsannscccsancensecasseeenes 1-200
`A, McCree - Texas Instruments, USA, K. Truong - Atlanta Signal Processors, Inc, USA, E. George -“Texas Instruments, USA
`T. Barnwell - Atlanta Signal Processors, Inc., USA, V. Viswanathan - Texas Instruments, USA
`
`Harmonic- Stochastic eXcitation (HSX) Speech Coding Below 4Kbits ...,....ss0sssseeessssueeeee acsusneecdenscsnenscerssseascesenssscccel~204
`C. Laflamme, R. Salami, R. Matmti, J. Adoul - University of Sherbrooke, Canada
`
`A High Quality MBE-LPC-FE Speech Coderat 2.4kbps and 1.2kbps_.....s.sseeeseee dae igavsasedavaceeswonssaccrsncasscossesersceaerssensl-20S
`T. Wang, K. Tang, C. Feng - Tsinghua University, China
`
`A Low-Complexity Waveform Interpolation Coder .......scessccserseressesenanes cannes TURPEAEENUG DOR eEHugNarNReAeKancrdesesanenveqeenqayarenel-212
`W. Kleijn, ¥. Shoham, D. Sen, R. Hagen - AT&T Bell Laboratories, USA
`
`Mixed-Domain Codingof Speech at 3 KDpS ...csssssseeseseereeess Sbencen head enadseadeeaveccaadvensdaansbbasascieebeacdiebedidudbedeacstebedss eeel-216
`J. De Martin - Politecnico di Torino, Italy
`A. Gersho - University of California at Santa Barbara, USA
`
`Source Driven/Variable Bit Rate Protoype Interpolation Coding ......ssscssserevesssenessoneeenens deneteveuensanteennsveasennoaretesttensel-220
`C. Xydeas, B. Cao - University ofManchester, UK
`
`A New Approachto Very Low-Rate Speech Coding Using Temporal Decomposition ..........cssssssceeeecsessceneeerneseeseesssesees1-224
`S. Ghaemmaghami, M. Deriche - Queensland University of Technology, Australia
`
`A Variable Frame Pitch Estimator and Test Results ....,......scscvessossscsesenercsnecceceecseecnenseeceseanercessssdeesenseeeseaneretesseecees1-228
`X. Qian, R. Kumaresan - University ofRhede Island, USA
`
`Robust Method of Measurement of Fundamental Frequency by ACLOS - AutoCorrelation of Log Spectrum -
`N. Kunieda, T. Shimamura, J. Suzuki - Satiama University, Japan
`
`........ssseesee0J-232
`
`...........sccessseseeessneeesensreenensnneverensrepapereeenencvaneusnaenenenetstanseeseeeseereneneneene1-236
`Lag-Indexed VQ for Pitch Filter Coding
`5. McClellan - University ofAlabama-Birmingham, USA
`J. Gibson - Texas A&M University, USA
`
`xii
`
`Page 19 of 72
`
`Page 19 of 72
`
`

`

`Volume I
`[aneeeaeee
`SP7 Wideband Coding and Emerging Techniques
`Embedded Algebraic Vector Quantizer (EAVQ) with Application to Wideband Speech Coding «......ccssssseeseseeeseeseetsecsees 1-240
`M. Xie, J. Adoul - University of Sherbrooke, Canada
`
`The Two-Dimensional Discrete Cosine Transform Applied to Speech Data ..........ccseccecseenseerennterstenessesssssussesenessceweneos 1-244
`L. Baghai-Ravary, S. Beet, M. Tokhi - University of Sheffield, UK
`
`Real-Time High Accurate Cell Loss Recovery Technique For Speech Over ATM Networks
`K. Matsumoto - NTT LSI Laboratories, Japan
`
`.....sssssssecsessseseeesssesssneeeeeesees1-248
`
`Predictive Fractal Interpolation Mapping: Differential Speech Coding at Low Bit Rates ........csssccssssesssseessssssreorssonsaeses1-251
`Z. Wang - University of Waterloo, Canada
`
` ..........:scsecessrseenensneeesasacessessccensscreneeesusseseuseeesanaaes1-255
`16kbit/s Wideband Speech Coding Based on Unequal Subbands
`J. Paulus, J. Schmitzler - IND, Aachen University of Technology, Germany
`
`Low Delay IIR QMFBankswith High Perceptive Quality for Speech Processing
`T. Kleinmann, A. Lacroix - University of Frankfurt, Germany
`
` .......sseressssssesssessessterescescssnensececeeeaaen1-259
`
`Demodulators for AM-FM Models of Speech Signals: A Comparison—...ssscserssesscceeerseensssseneneseseneaeenccesansaneunensaucunnsens 1-263
`S. Lu, P. Doerschuk - Purdue University, USA
`
`Synthesis and Coding of Continuous Speech with the Nonlinear Oscillator Model
`G. Kubin - Vienna University of Technology, Austria
`
`........-.ccsssescsssssssersessensessseennsnenenenseas1-267
`
`Variable Frame Rate Parameter Encoding via Adaptive Frame Selection Using Dynamic Programming—..ssssssrsssseseeessnees I-271
`E. George, A. McCree, V. Viswanathan - Texas Instruments, Inc., USA
`
`Transform Predictive Coding of Wideband Speech Signals ............sssccccsssscssssersscaserseccessesecsarsveasareceseasssensassseeseneaee-275
`J. Chen - AT&T Bell Labs, USA
`D. Wang - Georgia Institute af Technology, USA
`
`SP8__Topic Identification and Spoken Information Retrieval
`A System for Unrestricted Topic Retrieval From Radio News Broadcasts—....sssssesssereseeevsnnsssnerseeserensoecsscsooeasenseeeas eres1-279
`D. James - Union Bank of Switzerland, Switzerland
`
`Automated Generation of N-Best Pronunciations of Proper NOUNS—...ssecccerereeceensssceeeenseeeesseeveeeasseeaenererensmenreenennseanee1-283
`N. Deshmukh, M. Weber, J. Picone - Mississippi State University, USA
`
`An Efficient Voice Retrieval System for Very-Large-Vocabulary Chinese Textual Databases with a Clustered Language Model «1-287
`S. Lin - National Taiwan University, ROC
`L. Chien, K. Chen, L. Lee - Academia Sinica, ROC
`
`Concept-based Phrase Spotting Approach for Spontaneous Speech Understanding—.......sssecereeereressreeneeeeeeeesereeaenenenens1-291
`T. Kawahara, N. Kitaoka, S. Doshita - Kyoto University, Japan
`
`A Dictionary-Based Method for Determining Topics in Text and Transcribed Speech I-295
`P. Schone, D. Nelson - DepartmentofDefense, USA
`
`Keyword Spotting for Video Soundtrack Indexing ....ccssssseesersrrovserserssnrvenssecenttcenenseseaanensgeenseseenaaceseceusequcacauseaenees1-299
`P. Gelin, C. Wellekens - Institut Eurecom, France
`
`Improvements in Switchboard Recognition and Topic Identification ..........1..:ssccrereessenenesneseanensnseemneserenenassauamaannns sree1-303
`B. Peskin, S, Connolly, L. Gillick, 5. Lowe, D. McAllaster, V. Nagesha, P. van Mulbregt, S. Wegmann - Dragon Systems, Inc., USA
`
`Statistical Models for Topic Identification Using Phoneme Substrimgs
`J. Wright - University of Bristol, UK
`M. Carey, E. Parris - ENSIGMA Limited, UK
`
`..........scccosessesesseeeresesenneeseeanaaaesenueaaeeeeaneanens +«+.1-307
`
`.....1.:.seesseuee
`Robust Talker-Independent Audio Document Retrieval
`G. Jones, J. Foote, K. Spark Jones, S.