UNITED STATES PATENT AND TRADEMARK OFFICE
`WASHINGTON, D.C.
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`WAVES AUDIO, LTD.
`Petitioner
`
`v.
`
`ANDREA ELECTRONICS CORPORATION
`Patent Owner
`
`
`
`
`
`
`
`Case: IPR2016-00459
`
`Patent 6,363,345
`
`
`
`DECLARATION OF BERTRAND M. HOCHWALD IN SUPPORT OF
`PETITION FOR INTER PARTIES REVIEW OF US PATENT 6,363,345
`
`
`
`1
`
`
`
`
`
`WAVES345_1017-0001
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`WASHINGTON, D.C.
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`WAVES AUDIO, LTD.
`Petitioner
`
`v.
`
`ANDREA ELECTRONICS CORPORATION
`Patent Owner
`
`
`
`
`
`
`
`Case: IPR2016-00459
`
`Patent 6,363,345
`
`
`
`DECLARATION OF BERTRAND M. HOCHWALD IN SUPPORT OF
`PETITION FOR INTER PARTIES REVIEW OF US PATENT 6,363,345
`
`
`
`1
`
`
`
`
`
`WAVES345_1017-0001
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`I.
`
`INTRODUCTION
`
`1.
`
`I, Bertrand M. Hochwald, declare the following to be true and correct
`
`to the best of my knowledge. Except where otherwise indicated, I have personal
`
`knowledge of the facts below and, if called to do so, would and could testify
`
`regarding the matters stated in this report. I make this Declaration in support of inter
`
`partes review of US Patent 6,363,345.
`
`2.
`
`I have been retained by Denko Lauff LLP on behalf of Waves Audio,
`
`Ltd.
`
`3.
`
`I have been asked to provide my technical review, analysis, insights,
`
`and opinions on the materials I have reviewed in this case related to the ‘345 patent,
`
`including the references that form the basis of the grounds of rejection.
`
`II. QUALIFICATIONS
`
`A. Educational Background
`
`4.
`
`In 1995 I received a Ph.D. in Electrical Engineering, and in 1993 an
`
`M.A. in Statistics from Yale University. My primary area of study was Statistical
`
`Signal Processing. I received an M.S. in Electrical Engineering from Duke
`
`University in 1986, and a B.S. in Engineering from Swarthmore College in 1984.
`
`
`
`
`
`
`
`2
`
`WAVES345_1017-0002
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`B. Career History
`
`5.
`
`I have twenty years of combined industry and academic experience in
`
`the research and design of systems for signal processing, and wireless and wireline
`
`communications.
`
`6. My most recent appointment, starting in 2011, is with the University of
`
`Notre Dame, where I am currently a Freimann Chaired Professor of Electrical
`
`Engineering. I teach both graduate and undergraduate classes in Communication
`
`Systems and in Signals and Systems, where the emphasis is on the processing of
`
`analog and digital signals. My primary areas of research include communication
`
`systems, radiofrequency circuits, and signal design and processing. I advise
`
`graduate students who are attaining Ph.D. degrees through research and coursework.
`
`7.
`
`Prior to Notre Dame, I worked from 2005-2010 at Beceem
`
`Communications, a cellular wireless communication chipset start-up company in
`
`Santa Clara, California, where I was Chief Scientist and Vice President of Systems
`
`Engineering. I was an integral part of the chipset development team. Beceem was
`
`bought by Broadcom Corporation in 2010 and no longer exists as a separate
`
`company.
`
`8.
`
`Prior to Beceem, I worked from 1996-2005 at Lucent Bell Laboratories
`
`in New Jersey, where I was as a Distinguished Member of the Technical Staff doing
`
`research into communications systems and multiple-antenna systems. This resulted
`
`
`
`3
`
`WAVES345_1017-0003
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`in many patents and publications across a variety of areas in communication theory,
`
`information theory, and circuit design.
`
`9.
`
`Prior to Bell Laboratories, I was a Visiting Assistant Professor at the
`
`University of Illinois in Urbana-Champaign during the 1995-1996 school year,
`
`where I worked on a broad range of research topics related to signal processing and
`
`communications.
`
`10. Prior to completing my Ph.D., during 1986-1989 I worked at the
`
`Department of Defense as a system engineer for signal processing and wireless
`
`communication systems.
`
`C.
`
`Publications
`
`11. As indicated in detail in my CV (attached and incorporated as Exhibit
`
`1), I have published approximately 95 articles in scholarly journals, many of them
`
`within the journals of the Institute of Electrical and Electronic Engineers (IEEE),
`
`one of the premier societies for electrical engineers. I have 37 granted patents in a
`
`variety of areas related to communication and signal processing systems. I have
`
`been an invited and plenary speaker at several international conferences throughout
`
`the world and have received awards and recognition for my research and
`
`publications.
`
`
`
`
`
`
`
`4
`
`WAVES345_1017-0004
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`D. Other relevant qualifications
`
`12.
`
`In addition to my academic and practical experience, I have worked as
`
`an expert in the areas of communication and signal processing systems, as detailed
`
`in Exhibit 1. I have had experience drafting and successfully prosecuting my own
`
`patents, and have worked with other experts in signal processing systems as a co-
`
`inventor and co-author.
`
`13.
`
`I am being compensated for my time at the rate of $500 per hour for my
`
`work in connection with this matter. This compensation is not dependent in any way
`
`on the contents of this Declaration, the substance of any further opinions or
`
`testimony that I may provide or the ultimate outcome of this matter.
`
`III. MATERIALS REVIEWED
`
`14.
`
`In formulating my opinions, I have relied on my knowledge and
`
`experience in the field of signal and audio processing, and on the documents and
`
`information described below. I have reviewed the following materials:
`
`•
`
`•
`
`U.S. Patent No. 6,363,345 (the ‘345 patent) and its file history;
`
`Sources listed in Exhibit 2.
`
`IV. LEGAL PRINCIPLES
`
`15.
`
`I will not offer opinions of law because I am not an attorney. However,
`
`I have been informed of several principles of patent law, which I used in formulating
`
`my opinions.
`
`
`
`5
`
`WAVES345_1017-0005
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`16.
`
`In formulating my opinion, I understand that I must use a “broadest
`
`reasonable interpretation standard” for interpreting the claims in an inter partes
`
`Review. Further, there is no presumption of validity of the claims.
`
`A. Anticipation
`
`17.
`
`I understand that each patent claim is considered separately for
`
`purposes of invalidity. I have been informed that a patent claim is invalid as
`
`“anticipated” if each and every feature of the claim is found, expressly or inherently,
`
`in a single prior art reference or product arranged as recited in a claim. Claim
`
`limitations that are not expressly found in a prior art reference are inherent if the
`
`prior art necessarily functions in accordance with, or includes, the claim limitations.
`
`It is acceptable to examine evidence outside the prior art reference (extrinsic
`
`evidence) in determining whether a feature, while not expressly discussed in the
`
`reference, is necessarily present in it.
`
`B. Obviousness
`
`18.
`
`I have been informed that the combination of familiar elements
`
`according to known methods is likely to be obvious when it does no more than yield
`
`predictable results. In other words, when a claim simply arranges prior art elements
`
`with each performing the same function it had been known to perform and yields no
`
`more than one would expect from such an arrangement, then such a combination is
`
`obvious. Moreover, when a patent claims a structure already known in the prior art
`
`
`
`6
`
`WAVES345_1017-0006
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`that is altered by the mere substitution of one element for another known in the field,
`
`the combination is likely to be obvious unless the combination yields and
`
`unpredictable result.
`
`19.
`
`I have been informed that when a work is available in one field of
`
`endeavor, design incentives and other market forces can prompt variations of it,
`
`either in the same field or a different one. If one of ordinary skill in the art can
`
`implement a predictable variation, such a variation is likely unpatentable. For the
`
`same reason, if a technique has been used to improve one device, and one of ordinary
`
`skill in the art would recognize that it would improve similar devices in the same
`
`way, using the technique is obvious unless its actual application is beyond his or her
`
`skill. One question to consider is whether the improvement is more than the
`
`predictable use of prior art elements according to their established functions.
`
`20.
`
`I am informed that it may often be necessary, in a validity analysis, to
`
`look to interrelated teachings of multiple patents; the effects of demands known to
`
`the design community or present in the marketplace; and the background knowledge
`
`possessed by one of ordinary skill in the art, all in order to determine whether there
`
`was an apparent reason to combine the known elements in the fashion claimed by
`
`the patent at issue.
`
`21.
`
`I am informed that a validity analysis need not seek out precise
`
`teachings directed to the specific subject matter of the challenged claim; it is
`
`
`
`7
`
`WAVES345_1017-0007
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`appropriate to take account of the inferences and creative steps that one of ordinary
`
`skill in the art would employ.
`
`22.
`
`I understand that a claim composed of several elements is not proved
`
`obvious merely by demonstrating that each element was, independently, known in
`
`the prior art. I understand that it can be important to identify a reason that would
`
`have prompted a person of ordinary skill in the relevant field to combine the
`
`elements in the way the claimed new invention does. Therefore, I am informed that,
`
`in determining whether a claimed combination is obvious, the correct question is
`
`whether the combination was obvious to one of ordinary skill in the art at the time
`
`the patent application was filed. I am told that one of the ways in which subject
`
`matter can be proved obvious is by noting that there existed at the time of invention
`
`a known problem for which there was an obvious solution encompassed by the
`
`patent’s claims. I understand that any need or problem known in the field of endeavor
`
`at the time of alleged invention and addressed by the patent can provide a reason for
`
`combining the elements in the manner claimed.
`
`23.
`
`I am informed that one should not assume that a person of ordinary skill
`
`in the art attempting to solve a problem will be led only to those elements of prior
`
`art designed to solve the same problem. Instead, I understand that since familiar
`
`items may have obvious uses beyond their primary purposes, in many cases one of
`
`
`
`8
`
`WAVES345_1017-0008
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`ordinary skill in the art will be able to fit the teachings of multiple prior art references
`
`together like pieces of a puzzle.
`
`24.
`
`I am informed that, when there is a design need or market pressure to
`
`solve a problem and there are a finite number of identified, predictable solutions,
`
`one of ordinary skill in the art has good reason to pursue the known options within
`
`his or her technical grasp. If this leads to the anticipated success, then it is likely the
`
`product is not of innovation but of ordinary skill and common sense. I understand
`
`that, in such an instance, the fact that the combination was obvious to try may show
`
`that the combination is obvious.
`
`25.
`
`I am informed that, when determining whether a claimed combination
`
`is obvious, the correct analysis is not whether one of ordinary skill in the art, writing
`
`on a blank slate, would have chosen the particular combination of elements
`
`described in the claim. Instead, I understand the correct analysis considers whether
`
`one of ordinary skill, facing the wide range of needs created by developments in the
`
`field of endeavor, would have seen a benefit to selecting the combination claimed.
`
`C. Written Description
`
`26.
`
`I am informed that a patent must contain a written description of the
`
`invention to ensure that the patentee had possession of the claimed invention at the
`
`time of the application, i.e., that the patentee invented what is claimed. I am
`
`informed that there are two closely related requirements to meet the written
`
`
`
`9
`
`WAVES345_1017-0009
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`description requirement. First, it must describe the process or manner in the
`
`specification in a way that enables one of ordinary skill in the art to make and use
`
`the invention without undue experimentation. Second, the disclosure must describe
`
`the invention in a manner that one of ordinary skill in the art would know that the
`
`patentee had possession of all of the claimed invention at the time of the application.
`
`27.
`
`I am informed that a recitation of how to make and use the invention
`
`across the full breadth of the claim is ordinarily sufficient to demonstrate that the
`
`inventor possesses the full scope of the invention. I am also informed that a claim
`
`will not be invalidated for lack of written description grounds simply because the
`
`embodiments of the specification do not contain examples explicitly covering the
`
`full scope of the claim language. However, I am also informed that there are
`
`situations in which a specification that describes a species might fail to adequately
`
`describe a claim to the genus. I am informed that a single embodiment would
`
`support such a generic claim only if the specification would reasonably convey to a
`
`person skilled in the art that the inventor had possession of the full scope of the
`
`claimed invention at the time of filing.
`
`28.
`
`I also understand that a patent must conclude with one or more claims
`
`particularly pointing out and distinctly claiming the subject matter which the
`
`applicant regards as his invention. To satisfy this definiteness requirement, the
`
`claims of a patent must adequately perform their function of notifying the public of
`
`
`
`10
`
`WAVES345_1017-00010
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`the scope of the patentee’s right to exclude. This standard is only met when the
`
`claims clearly distinguish what is claimed from what went before in the art and
`
`clearly circumscribe what is foreclosed from future enterprise.
`
`29. Consequently, the claims must be sufficiently precise to permit a
`
`potential competitor to objectively determine whether or not his actions would
`
`infringe.
`
`V. RESERVATION OF RIGHTS
`
`I reserve the right to amend and/or supplement this declaration if new
`
`evidence, arguments, or testimony is presented, for example, in discovery for
`
`this IPR.
`
`VI. GENERAL FIELD OF ART; A PERSON HAVING ORDINARY
`SKILL IN THE ART
`30.
`
`I understand from counsel that I should consider what is called the
`
`intrinsic evidence which includes the specification, including the claims and figures
`
`and text, and the file history as well as materials that have been incorporated into the
`
`specification from the perspective of a person having ordinary skill in the art at the
`
`time of the invention.
`
`31. Based on my experience and research in the field and knowledge of
`
`colleagues and co-workers and my experience as a professor of students ranging
`
`from candidates for the B.S. degree to Ph.D. candidates in electrical engineering, I
`
`
`
`11
`
`WAVES345_1017-00011
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`would characterize a person having ordinary skill in the art at the time of the
`
`invention (“PHOSITA”) as having either of the following qualifications:
`
`a)
`
`A Master’s degree in electrical engineering, computer science or
`
`physics or a related degree and familiarity and experience with
`
`digital signal processing; or
`
`b)
`
`A Bachelor’s degree in electrical engineering, computer science or
`
`physics or a related field with at least two years of industry or
`
`academic research experience in digital signal processing.
`
`VII. THE ‘345 PATENT
`
`A.
`
`Perspective
`
`32. The ‘345 patent, filed on February 18, 1999, is directed to a digital
`
`signal processing system for cancelling noise, specifically within an audio speech
`
`signal. The system uses a threshold detector to set a threshold for each frequency
`
`bin created by a frequency spectrum generator and compare the magnitude of each
`
`of the frequency bin to their corresponding threshold to detect the position of noise
`
`elements within speech and noise. The threshold is set according to current and
`
`future minimum values. The intent is to identify noise components embedded within
`
`the speech.
`
`33. Once the noise is identified it is removed using spectral subtraction
`
`using a filter multiplication. Residual noise reduction is used after noise removal.
`
`
`
`12
`
`WAVES345_1017-00012
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`34. The spectral subtraction method is described in a 1979 paper by Steve
`
`Boll, which is included as a reference in the Information Disclosure of the ‘345
`
`patent. Boll suggests subtracting the noise spectrum from the combined speech-
`
`plus-noise spectrum on a bin by bin basis to get a noise-reduced representation of
`
`speech.
`
` Boll also suggests methods for magnitude averaging, half-wave
`
`rectification, residual noise reduction, and additional signal attenuation during non-
`
`speech activity as further methods to reduce noise in speech and audio signals. A
`
`key requirement of Boll’s spectral subtraction method is an estimate of the noise
`
`spectrum. As suggested by Boll, this noise spectrum can be captured as an average
`
`during non-speech activity, provided good speech activity detection methods are
`
`available.
`
`35. Many publications and inventions since the Boll paper have focused on
`
`methods to estimate the noise spectrum, and thus enable spectral subtraction. Many
`
`publications have also focused on the other aspects of Boll’s magnitude averaging,
`
`half-wave rectification, and residual noise reduction.
`
`36. Generalizations of this spectral subtraction algorithm also exist. For
`
`example, Lim and Oppenheim, “Enhancement and Bandwidth Compression of
`
`Noisy Speech,” Proceedings of the IEEE, vol. 67, no. 12, pp. 1586-1604, December
`
`1979 disclose in Equation (12) of their work a spectral subtraction method as:
`
`(cid:3627)(cid:1845)(cid:4632)(cid:3050)((cid:2033))(cid:3627)(cid:3028)=|(cid:1851)(cid:3050)((cid:2033))|(cid:3028)−(cid:1831)[|(cid:1830)(cid:3050)((cid:2033))|(cid:3028)]
`
`
`
`13
`
`WAVES345_1017-00013
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`where (cid:1851)(cid:3050)((cid:2033)) is the input frequency bin, (cid:1830)(cid:3050)((cid:2033)) is the estimated noise frequency bin,
`and (cid:1845)(cid:4632)(cid:3050)((cid:2033)) is the estimate of the speech signal frequency bin and the E[] operation
`chosen between one and two. Setting (cid:1853)=1 gives the spectral subtraction method
`in Boll where the magnitude of the frequency bins is subtracted. Setting (cid:1853)=2 gives
`
`frequency. The constant “a” represents an exponent of the user’s choice, normally
`
`is expected value (usually obtained as a time average). The symbol “ω” denotes
`
`spectral subtraction where the power (rather than magnitude) of the frequency bins
`
`is subtracted.
`
`37.
`
`In view of Lim and Oppenheim and the above equation, a PHOSITA
`
`algorithms and expressions involving power ((cid:1853)=2) into those involving magnitude
`((cid:1853)=1). Similarly, a PHOSITA would be able to translate expressions involving
`
`would readily be able to translate spectral subtraction and noise estimation
`
`magnitude into those involving power.
`
`38. Whatever the choice for a, the essential considerations of spectral
`
`subtraction remain the same. In particular, the issue of how to estimate the noise
`
`frequency bin magnitudes |(cid:1830)(cid:3050)((cid:2033))| is a very important part of the process of spectral
`subtraction. Because spectral subtraction involving (cid:1853)=1 and (cid:1853)=2 was disclosed
`
`in the 1970’s, and the ‘345 patent was filed in 1999, there are two intervening
`
`decades’ worth of research and development work for a PHOSITA to draw upon, as
`
`
`
`14
`
`WAVES345_1017-00014
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`prior art, to provide methods of estimating the noise frequency elements or bins to
`
`enable spectral subtraction.
`
`39. The ‘345 Patent considers spectral subtraction with (cid:1853)=1 in the
`(cid:1853)=2 is available as potential anticipatory evidence for the ‘345 Patent Claims since
`
`Specification and Claims. However, prior art that employs spectral subtraction with
`
`the problem of detecting and estimating the noise frequency bins is the core subject
`
`of the patent, and this problem exists for all values of a.
`
`B. Notational Convention
`
`40.
`
`In the sections that follow, I discuss the Claims of the ‘345 patent in
`
`relation to certain prior art references. It is my conclusion that the prior art either
`
`anticipates or renders obvious the ‘345 Claims. I use italic emphasis in my
`
`discussion when referring to specific terms in the Claim language of the ‘345 patent.
`
`I use “quotations” when referring to specific terms in the prior art and references.
`
`C. The Claims of the ‘345 Patent
`
`The Claims discussed in this Declaration include Claims 1-3, 12-14, 17, 21,
`
`23, 25, 38, and 47.
`
`41.
`
`I reproduce Claim 1 for ease of reference:
`
`Claim 1. An apparatus for canceling noise, comprising:
`an input for inputting an audio signal which includes a noise signal;
`
`
`
`15
`
`WAVES345_1017-00015
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`a frequency spectrum generator for generating the frequency spectrum of
`said audio signal thereby generating frequency bins of said audio
`signal;
`a threshold detector for setting a threshold for each frequency bin using a
`noise estimation process and for detecting for each frequency bin
`whether the magnitude of the frequency bin is less than the
`corresponding threshold, thereby detecting the position of noise
`elements for each frequency bin.
`42. A PHOSITA at the time of filing of this patent would interpret this
`
`claim as being directed to an apparatus that is intended to separate an audio signal
`
`into frequency bins and then setting a threshold for each frequency bin to determine
`
`if there is noise in that frequency bin. The determination of noise is made if the
`
`magnitude of the frequency bin is less than the threshold for that bin.
`
`43. Claim 2 depends on Claim 1 and adds plurality and position of non-
`
`speech points.
`
`44. Claim 3 depends on Claim 2 and adds application to continuous speech
`
`segment.
`
`45. Claim 12 depends on Claim 1 and adds an averaging unit for
`
`determining level of comparison against a threshold. The Specification discloses
`
`that the term averaging unit as used in the ‘345 patent refers to an operation such as
`
`depicted in block 310 in the figures, which uses a recursive filter to estimate the
`
`noise level. The ‘345 Specification col. 6, line 51 refers to 310 as an averaging unit.
`
`
`
`16
`
`WAVES345_1017-00016
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`46. Claim 13 depends on Claim 1 and includes a subtractor for subtracting
`
`the noise spectrum from the audio signal.
`
`47. Claim 14 depends on 13 and is a variation on the subtractor using a
`
`filter multiplication.
`
`48. Claim 17 depends on Claim 13 and includes a residual noise processor
`
`for removing residual noise after subtraction.
`
`49. Claim 21 depends on Claim 1 and provides an estimator for the
`
`magnitude.
`
`50. Claim 23 depends on Claim 21 and provide an estimator and smoothing
`
`unit for estimating the magnitude of each frequency bin. The term smoothing unit
`
`in the ‘345 patent is not clearly defined in the Specification. The operation of
`
`smoothing, however, refers to an operation such as depicted in block 208, which
`
`shows a weighted-averager. The term smoothing is also used to refer to block 206,
`
`either alone or in combination with 208. However, even though the term smoothing
`
`unit is used in Claim 23 and the term averaging unit is used in Claim 12, the
`
`distinction between smoothing and averaging is not made clear in the Specification
`
`since the terms smoothing and averaging are used interchangeably, for example, in
`
`col. 5 lines 45-60 of the ‘345 Specification. In fact, the term exponential average is
`
`used to describe both blocks 208 and 310, and 310 is identified by the Specification
`
`as an averaging unit (see my discussion of Claim 12 above). .
`
`
`
`17
`
`WAVES345_1017-00017
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`51. Claim 25 depends on Claim 1 and includes an adaptive array
`
`comprising a plurality of microphones.
`
`52. Claim 38 is a method claim incorporating elements from Claims 1 and
`
`13. It is reproduced here:
`
`Claim 38. A method for driving a computer processor for generating a
`noise canceling signal for canceling noise from an audio signal
`representing audible sound including a noise signal representing
`audible noise, said method comprising the steps of:
`inputting said audio signal which includes said noise signal;
`generating the frequency spectrum of said audio signal thereby generating
`frequency bins of said audio signal;
`setting a threshold for each frequency bin using a noise estimation
`process;
`detecting for each frequency bin whether the magnitude of the frequency
`bin is less than the corresponding threshold, thereby detecting the
`position of noise elements for each frequency bin; and
`subtracting said noise elements detected in said step of detecting from said
`audio signal to produce an audio signal representing said audible
`sound substantially without said audible noise.
`53. A PHOSITA would interpret this as a method for cancelling noise from
`
`an audio signal, wherein the audio signal is divided into frequency bins, thresholds
`
`are set for each bin, and using the threshold to determine whether noise is in that bin.
`
`The noise that is detected is then subtracted from the input audio signal to generate
`
`a noise-canceled audio signal.
`
`
`
`18
`
`WAVES345_1017-00018
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`54. Note that Claim 38 mentions a method of generating a noise canceling
`
`signal, in the introductory sentence, but this canceling signal is not mentioned in the
`
`remainder of the claim or in its dependents. Instead, the fourth clause of this claim
`
`defines noise elements which are ultimately subtracted from the audio signal, and
`
`hence canceled from the audio signal. I therefore conclude that the term noise
`
`canceling signal actually refers to the noise elements that are subtracted from the
`
`audio signal and attach no further significance to the term noise canceling signal.
`
`55. Claim 47 depends on Claim 38 and includes the step of reducing the
`
`residual noise remaining after subtraction.
`
`VIII. COMPARISON OF THE ’345 PATENT TO THE PRIOR ART
`
`A. Hirsch (1995)
`
`56.
`
`In my opinion, Hirsch anticipates ‘345 Claims 1-3, 12, 13, 21, 23, and
`
`38.
`
`57. Hirsch describes a technique for estimating the noise spectrum in
`
`continuous audio or speech signals. The goal in Hirsch is to estimate the noise
`
`spectrum, to allow an application of a variant on Boll’s spectrum subtraction method.
`
`Anticipation of ‘345 Patent
`
`58. Claim 1: Equation (1) on pg. 153 in Hirsch denotes an estimator of
`
`spectral magnitude of the noise, where the index “i” indicates the frequency bin.
`
`
`
`19
`
`WAVES345_1017-00019
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`Hence there is a frequency spectrum generator for generating the frequency
`
`spectrum of said audio.
`
`59. The input to this equation is “Xi(k)” which can contain speech and noise
`
`and is therefore the input to the estimator that is an audio signal which includes a
`
`noise signal. As described in the text immediately preceding Equation (1), the value
`
`“Xi” corresponds “spectral magnitude” and hence corresponds to magnitude of the
`
`frequency bin.
`
`60. As described in the paragraphs immediately following Equation (1),
`
`this input is compared to a threshold to determine if the input is noise or speech. If
`
`the threshold is not exceeded then Equation (1) is used; if the threshold is exceeded
`
`then Equation (1) is not used. Hence there is a threshold detector. This threshold
`
`detector is used to detect the position of noise elements for each frequency bin.
`
`Consequently all of the limitations of Claim 1 are found in Hirsch.
`
`61. Claim 2: a plurality of non-speech data points is disclosed in Equation
`
`(1) as the equation includes a time index “k” that is allowed to run indefinitely. The
`
`time index indicates the position. Therefore Claim 2 of the ‘345 Patent is
`
`anticipated.
`
`62. Claim 3: continuous speech is present in the Hirsch reference where it
`
`describes how explicit speech pause detection is not needed, and “continuous
`
`
`
`20
`
`WAVES345_1017-00020
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`spectral subtraction” is used in his algorithm. Therefore, Claim 3 of the ‘345 Patent
`
`is anticipated.
`
`63. Claim 12: An averaging unit is disclosed in equation (1) of Hirsch,
`
`which includes a weighted averaging term for determining the level of the noise in
`
`the frequency bin. This equation is: (cid:1840)(cid:3553)(cid:3036)((cid:1863))=(1−⍺)(cid:1850)(cid:3036)((cid:1863))+⍺(cid:1840)(cid:3553)(cid:3036)((cid:1863)−1), where
`“1-⍺” determines how much the value of “Xi” influences the averaging. Therefore,
`
`time is indexed by “k” and frequency bin is determined by “i”. The weighting factor
`
`‘345 Claim 12 is anticipated.
`
`64. Claim 13: A subtractor for subtracting said noise elements is likewise
`
`disclosed in the Hirsch reference. For example, 153, col. 2, refers to “continuous
`
`spectral subtraction”. Page 154, col. 2, refers to “nonlinear spectral subtraction”.
`
`These disclosures of “continuous spectral subtraction” demonstrate that Claim 13 of
`
`the ’345 Patent is anticipated.
`
`65. Claim 21: An estimator for estimating a magnitude is similarly
`
`disclosed by the Hirsch reference as Equation (1) in Hirsch defines “Xi(k)” to be a
`
`“spectral magnitude”. Thus, Claim 21 of the ’345 Patent is anticipated by Hirsch.
`
`66. Claim 23: A smoothing unit is disclosed in Equation (1) of the Hirsch
`
`reference as it includes first-order recursive filter to smooth spectral magnitude
`
`values. Hirsch on page 154, second paragraph says “The estimated values for the
`
`
`
`21
`
`WAVES345_1017-00021
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`noise magnitude are smoothed versus time to eliminate rarely occurring spikes.”
`
`Hence, ‘345 Claim 23 is anticipated.
`
`67. Claim 38: Claim 38 is a recitation of Claim 13, which in turn depends
`
`from apparatus Claim 1, in the form of a method claim. As a result, Hirsch
`
`anticipates Claim 38 of the ’345 Patent for the same reasons that Claim 13 is
`
`anticipated.
`
`B. Higgins (1998)
`
`68.
`
`In my opinion, Higgins anticipates ‘345 Claims 1-3, 13, and 38.
`
`69. Higgins details an apparatus/method for noise suppression for speech
`
`and speaker recognizers. The method samples the incoming noisy voice signal,
`
`generates a spectral representation, creates a noise threshold, and suppresses the
`
`noise with spectral subtraction.
`
`Anticipation of ‘345 Patent
`
`70. Claim 1: Figure 1 in Higgins discloses a processing system wherein the
`
`input is a microphone, which is an input for inputting an audio signal which includes
`
`a noise signal.
`
`71. Figure 2A in Higgins shows a frequency spectrum generator for
`
`generating the frequency spectrum of the audio signal. This spectrum generator is
`
`used to decompose the input signal into “an N-point spectral sample” (Higgins col.
`
`3, line 55)
`
`
`
`22
`
`WAVES345_1017-00022
`
`Petitioner Waves Audio Ltd. 345 - Ex. 1017
`
`
`
`72. Higgins col. 3 line 56 describes a threshold detector: “detecting a peak
`
`amplitude of the histogram which corresponds to a noise threshold Nf associated
`
`with the particular frequency. Col. 9 line 16 then describes that the “channel
`
`frequency response” is them computed in module 90 by determining the geometric
`
`mean over the utterance of those magnitudes at frequency f that exceed the noise
`
`floor Nf. Hence the position of noise elements for each frequency bin is detected.
`
`73. Hence, the limitations of Claim 1 of ‘345 are met and the claim is
`
`therefore anticipcated.
`
`74. Claim 2: According to Higgins col. 9, line 1: “This requires steps 50
`
`and 60 (windowing and FFT processing) to be performed for each frame t of sampled
`
`data and module 70 (rectangular to polar conversion) to be performed for each frame
`
`t and each frequency f.” This shows that the threshold detector detects the position
`
`of a plurality of non-speech data points for said frequency bins, and Claim 2 of the
`
`‘345 Patent is anticipated.
`
`75. Claim 3: According to Higgins col. 8, line 65: “Referring now to Fig.
`
`3 in conjunction with Figs. 2A and B, at a first pass the magnitudes computed by
`
`module
Accessing this document will incur an additional charge of $.
After purchase, you can access this document again without charge.
Accept $ Charge
Still Working On It
This document is taking longer than usual to download. This can happen if we need to contact the court directly to obtain the document and their servers are running slowly.
Give it another minute or two to complete, and then try the refresh button.
A few More Minutes ... Still Working
It can take up to 5 minutes for us to download a document if the court servers are running slowly.
Thank you for your continued patience.
This document could not be displayed.
We could not find this document within its docket. Please go back to the docket page and check the link. If that does not work, go back to the docket and refresh it to pull the newest information.
Your account does not support viewing this document.
You need a Paid Account to view this document. Click here to change your account type.
Your account does not support viewing this document.
Set your membership
status to view this document.
With a Docket Alarm membership, you'll
get a whole lot more, including:
- Up-to-date information for this case.
- Email alerts whenever there is an update.
- Full text search for other cases.
- Get email alerts whenever a new case matches your search.
One Moment Please
The filing “” is large (MB) and is being downloaded.
Please refresh this page in a few minutes to see if the filing has been downloaded. The filing will also be emailed to you when the download completes.
Your document is on its way!
If you do not receive the document in five minutes, contact support at support@docketalarm.com.
Sealed Document
We are unable to display this document, it may be under a court ordered seal.
If you have proper credentials to access the file, you may proceed directly to the court's system using your government issued username and password.
Access Government Site
