`__________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`__________________________________________________________________
`
`
`
`GOOGLE LLC,
`
`Petitioner,
`
`v.
`
`VOCALIFE LLC,
`
`Patent Owner.
`
`Patent No. RE47,049
`Filing Date: October 14, 2016
`Issue Date: October 14, 2014
`Reissue Date: September 18, 2018
`
`Title: MICROPHONE ARRAY SYSTEM
`
`__________________________________________________________________
`
`PATENT OWNER’S RESPONSE
`
`Case No. IPR2022-00004
`__________________________________________________________________
`
`
`
`
`
`TABLE OF CONTENTS
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`Page(s)
`
`
`I.
`INTRODUCTION ........................................................................................... 1
`DR. QI “PETER” LI AND HIS COMPANIES............................................... 1
`II.
`III. LITIGATION HISTORY BETWEEN VOCALIFE AND GOOGLE ............ 2
`IV. THE DEPOSITION OF GOOGLE’S EXPERT, DR. JEFFREY
`VIPPERMAN .................................................................................................. 6
`THE LAW OF OBVIOUSNESS ................................................................... 10
`V.
`VI. THE PROPOSED JEONG COMBINATIONS DO NOT DISCLOSE
`OR RENDER OBVIOUS THE DIGITAL SIGNAL PROCESSOR
`LIMITATIONS, INCLUDING THE THREE RECITED UNITS
`INTEGRATED IN A SINGLE DIGITAL SIGNAL PROCESSOR ............ 12
`VII. JEONG DOES NOT MEET THE “DETERMINING A DELAY”
`LIMITATIONS .............................................................................................. 17
`VIII. JEONG DOES NOT RENDER OBVIOUS THE “PLURALITY OF
`CONFIGURATIONS” LIMITATION .......................................................... 19
`IX. THE PROPOSED JEONG COMBINATION DOES NOT
`DISCLOSE OR RENDER OBVIOUS THE “WHEREIN SAID
`DELAY IS REPRESENTED IN TERMS OF NUMBER OF
`SAMPLES” LIMITATION ........................................................................... 25
`THE PROPOSED JEONG COMBINATIONS DO NOT DISCLOSE
`OR RENDER OBVIOUS THE SOUND SOURCE
`LOCALIZATION UNIT TO ENABLE BEAMFORMING
`LIMITATIONS .............................................................................................. 27
`XI. THE PRIOR ART DOES NOT DISCLOSE OR RENDER
`OBVIOUS DEPENDENT CLAIMS 2-8, 10-19, 23-25, AND 27-29 .......... 31
`XII. CONCLUSION .............................................................................................. 33
`
`
`X.
`
`i
`
`
`
`
`
`TABLE OF AUTHORITIES
`
`IPR2022-00004
`PATENT NO. RE47,049
`
` Page(s)
`
`Cases
`Amazon.com, Inc. v. Vocalife, LLC,
`IPR2020-00864, Paper 22 (P.T.A.B. Oct. 28, 2020) ............................................ 4
`Amazon.com, Inc. v. Vocalife LLC,
`IPR2020-00864, Paper 26 (P.T.A.B. Jan. 22, 2021) ............................................ 3
`Apple, Inc. v. Contentguard Holdings, Inc.,
`IPR2015-00355, Paper 9 (P.T.A.B. June 26, 2015) ........................................... 12
`Broadcom Corp. v. Emulex Corp.,
`732 F.3d 1325 (Fed. Cir. 2013) .......................................................................... 12
`Graham v. John Deere Co. of Kan. City,
`383 U.S. 1 (1966) ................................................................................................ 11
`Intelligent Bio-Sys., Inc. v. Illumina Cambridge Ltd.,
`821 F.3d 1359 (Fed. Cir. 2016) .......................................................................... 11
`Los Angeles Biomedical Research Inst. At Harbor-UCLA Med. Ctr. v.
` Eli Lilly & Co., 849 F.3d 1049 (Fed. Cir. 2017) ............................................... 11
`Pers. Web Techs., LLC v. Apple, Inc.,
`848 F.3d 987 (Fed. Cir. 2017) ............................................................................ 11
`Stratoflex, Inc. v. Aeroquip Corp.,
`713 F.2d 1530 (Fed. Cir. 1983) .......................................................................... 11
`Vocalife LLC v. Bose Corporation,
`No. 2:21-cv-00128-JRG, Dkt. 38 (E.D. Tex. Oct. 13, 2021) ............................... 2
`Vocalife LLC v. Bose Corporation,
`No. 2:21-cv-00128-JRG, Dkt. 93 (E.D. Tex. Apr. 26, 2022) ............................... 3
`Vocalife LLC v. Harman Int'l Indus. Inc.,
`No. 2:21-cv-00123-JRG, Dkt. 61 (E.D. Tex. Sept. 10, 2021) .............................. 2
`
`
`
`ii
`
`
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`EXHIBITS
`
`Description of Document
`
`Decision Denying Institution in IPR2020-00864
`Claim Construction Order from Vocalife LLC v. Amazon.com, Inc.
`et al., USDC EDTX No. 2:19-CV-00123-JRG, Dkt. 83, dated April
`6, 2020.
`Joint Agreed Claim Construction Chart from Vocalife LLC v.
`Harman Becker International Industries, Inc., USDC EDTX No.
`2:21-CV-00123-JRG, Dkt. 70-1, dated Oct. 29, 2021.
`Transcript of the Deposition of Jeffrey S. Vipperman, Ph.D., taken
`July 12, 2022
`Declaration of Joseph C. McAlexander III
`
`
`
`
`
`
`
`
`Exhibit
`No.
`2001
`2002
`
`2003
`
`2004
`
`2005
`
`iii
`
`
`
`I.
`
`INTRODUCTION
`On April 15, 2022, the Board instituted a petition filed by Google LLC
`
`(“Petitioner” or “Google”) for inter partes review (“IPR”) of U.S. Patent No.
`
`RE47,049 (Ex. 1001, “the ’049 Patent”), authorizing review of Claims 1-35 (“the
`
`Challenged Claims”). Paper 10 at 2. Patent Owner Vocalife LLC (“Patent Owner”
`
`or “Vocalife”) files this response in opposition to the grounds set forth in the Petition
`
`as authorized for review by the Board.
`
`The Board made clear that its preliminary findings and conclusions were
`
`“based on the evidentiary record developed thus far” at the time of the Institution
`
`Decision and did not consider any testimony from Vocalife’s expert, Mr. Joseph
`
`McAlexander. Paper 10 at 2-3. The Board also did not consider the deposition
`
`testimony of Petitioner’s expert, Dr. Jeffrey Vipperman. Exhibit 2004.
`
`As set forth herein, in view of the evidence of record, Petitioner has not met
`
`its burden of providing any proposition of unpatentability by a preponderance of the
`
`evidence.
`
`II. DR. QI “PETER” LI AND HIS COMPANIES
`The ’049 Patent lists Dr. Qi “Peter” Li as one of two inventors. Dr. Li is also
`
`the owner of Vocalife and its sister company Li Creative Technologies.1 Dr. Li’s
`
`companies were started by former Bell Labs research scientists and have become
`
`
`1 See, e.g., https://licreative.com/ and https://crispmic.com/
`
`1
`
`
`
`
`recognized leaders in the technology fields of digital signal processing, speech
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`recognition, natural language processing, far-field microphone array technology, 3D
`
`audio and music, among others. Dr. Li was the winner of the International CES
`
`Innovations Design and Engineering Award and has been awarded numerous federal
`
`grants for research and development, including for numerous DARPA projects.
`
`Dr. Li’s companies provide new products, designs, services, and solutions for
`
`government and commercial enterprises,
`
`including
`
`smartphones, video
`
`conferencing, home automation, music, banking, and transportation industries. As
`
`part of these advancements, the ’049 Patent claims an invention related to far-field
`
`microphone array technology.
`
`III. LITIGATION HISTORY BETWEEN VOCALIFE AND
`Vocalife and Google are parties to Vocalife LLC v. Google LLC, Case No.
`
`2:21-cv-00124-JRG (E.D. Tex. Apr. 2, 2021) (“District Court Case”). The case was
`
`consolidated with Vocalife LLC v. Harman Int’l Indus. Inc., Case No. 2:21-cv-
`
`00123-JRG (E.D. Tex.) as the lead case. Defendant Harman International Industries
`
`Inc. took a license to the Vocalife patents. See Vocalife LLC v. Harman Int’l Indus.
`
`Inc., Case No. 2:21-cv-00123-JRG, Dkt. 61 (E.D. Tex. Sept. 10, 2021).2
`
`
`2 Defendants Bose Corporation and Sonos, Inc. have each taken a license to the
`Vocalife patents. See Vocalife LLC v. Bose Corporation, No. 2:21-cv-00128, Dkt.
`
`
`2
`
`
`
`Vocalife and Defendants Amazon.com, Inc. and Amazon.com, LLC were
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`
`
`parties to Vocalife LLC v. Amazon.com, Inc., Case No. 2:19-cv-00123-JRG (E.D.
`
`Tex. Apr. 16, 2019) (“Vocalife I”) which progressed to a jury trial before the
`
`Honorable Rodney Gilstrap. Vocalife I resulted in a final judgment entering a jury
`
`verdict of infringement and validity of claims 1 and 8 of the ’049. See Exhibit 2001.
`
`The jury entered a $5,000,000 verdict to be paid by Amazon. Id.
`
`During Vocalife I, Amazon filed a concurrent petition in IPR2020-00864
`
`which was denied institution. In its decision, the Board found “the merits are weak.”
`
`Amazon.com, Inc. v. Vocalife LLC, IPR2020-00864, Paper 26 at 3 (P.T.A.B. Jan. 22,
`
`2021).
`
`Many of the same limitations existing in both the ’049 Patent and ’371 Patent
`
`have already withstood the scrutiny of Amazon’s four trial law firms and have been
`
`found to be valid after a five-day trial on the merits, as well as the weak grounds in
`
`IPR2020-00864. The instant Petition simply re-packages many of the same old
`
`references that were previously relied on by Amazon in Vocalife I and subsequently
`
`dropped prior to trial.
`
`The Petition asserts TEN GROUNDS. Pet. at 4. All 10 Grounds are based on
`
`variations of the Jeong reference with one or more of the Van Trees, Briere, Buck-
`
`
`38 (E.D. Tex. Oct. 13, 2021); Vocalife LLC v. Bose Corporation, No. 2:21-cv-00128,
`Dkt. 93 (E.D. Tex. Apr. 26, 2022).
`
`3
`
`
`
`
`II, Kim, Yen, and Andrea references. The prior art references and obviousness
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`grounds cannot render the Challenged Claims obvious because they fail to disclose
`
`all of the elements required by the Challenged Claims. For example, the prior art
`
`references do not disclose or render obvious the Challenged Claims, particularly
`
`with respect to the following limitations: (1) the claimed digital signal processor
`
`with the three recited units integrated in a signal DSP; (2) determining a delay;
`
`(3) plurality of configurations; (4) delay is represented in terms of number of
`
`samples; and (5) sound source localization unit enables beamforming.
`
`First, Jeong does not teach any “digital signal processor” or the three recited
`
`units integrated in one single DSP, as required by the Challenged Claims. Similar to
`
`the deficiency in IPR2020-00864, where PTAB denied institution of a petition
`
`challenging the ’049 Patent finding that Petitioner failed to sufficiently show that the
`
`three recited units are “integrated in a digital signal processor.” Amazon v. Vocalife,
`
`LLC, IPR2020-00864, Paper 22 at 16 (P.T.A.B. Oct. 28, 2020). Briere does not cure
`
`Jeong’s deficiencies because Briere expressly admits that it was “impossible to build
`
`the original system on the DSP” and speculates that it would need to “reduce” the
`
`localization system. Briere at 3. The Petition’s failure to address these concessions
`
`of impossibility and inoperability are fatal to institution because they are direct
`
`evidence of teaching away from the claimed invention and direct evidence against
`
`any conclusory allegations of motivation to combine with Briere.
`
`4
`
`
`
`Second, Jeong does not teach the “determination of a delay” limitation, as
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`
`
`required by the Challenged Claims. Petitioner inconsistently maps both the
`
`“delaying unit” and the disclosed TDOA to this claim limitation. Such inconsistent
`
`mapping is fatal to the Petition. In addition, Petitioner’s expert concedes that the
`
`equation of the “delaying unit” disclosed in Jeong does not teach how to compute
`
`the delay. Further, Jeong’s TDOA is a “sound source location search method” and
`
`the TDOA Petitioner relies upon is not the same TDOA disclosed in the Challenged
`
`Claims.
`
`Third, Jeong and Van Trees do not render obvious the “plurality of
`
`configurations” limitation of the Challenged Claims where Petitioner concedes
`
`Jeong discloses a linear array configuration and Van Trees discloses a plurality of
`
`array configuration but fails to provide a motivation to combine these disclosures.
`
`Petitioner’s expert testified that each geometry would have its own specific delay
`
`and that each geometry would require different delays.
`
`Fourth, Jeong in view of Van Trees and Briere does not render obvious the
`
`“wherein said delay is represented in terms of number of samples” where Jeong
`
`discloses a time-based delay. There is no teaching in Jeong regarding delay
`
`represented in number of samples and Petitioner does not point to any disclosures of
`
`Van Trees that disclose delay represented in samples. Briere’s “sample frames” do
`
`not meet this limitation where Petitioner cannot show the “sample frames” are
`
`5
`
`
`
`
`representations for delay. Rather, Petitioner attempts to minimize the requirement of
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`the Challenged Claims that the delay must be represented in number of samples.
`
`Lastly, Jeong does not disclose or render obvious the limitations regarding the
`
`sound source
`
`localization unit
`
`to enable beamforming, where Petitioner
`
`inconsistently maps two different paths disclosed in Briere to meet the claimed
`
`“sound source localization unit” limitation, and the beamforming Petitioner points
`
`to is used to enable the finding of sources, the opposite of what the Challenged
`
`Claims require. Briere teaches a Sound Source Localization and Tracking system
`
`which uses a beamformer to find the position of a source and the tracking module to
`
`then follow the moving source. Briere further discloses that the sound source
`
`localization and tracking system is not used for speech recognition.
`
`IV. THE DEPOSITION OF GOOGLE’S EXPERT, DR. JEFFREY
`VIPPERMAN
`On July 12, 2022, Vocalife took the deposition of Google’s expert, Jeffrey S.
`
`Vipperman, Ph.D. See Ex. 2004. The deposition exposed numerous inconsistencies
`
`with certain positions set forth in the Petition. Such inconsistencies are identified in
`
`detail below.
`
`As a preliminary matter, Dr. Vipperman conceded that he did not look at the
`
`Petition at all in submitting his declaration.
`
`6
`
`
`
`
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`
`Ex. 2004, Vipperman Dep. Tr. at 20:9-19. Dr. Vipperman acknowledges that it may
`
`have “ma[d]e sense to have reviewed it.” Id.
`
`Further, Dr. Vipperman testified that “determination of delay” requires that
`
`each configuration have its own set of delays and that determination of delay for one
`
`configuration would not enable beamforming for other configurations.
`
`7
`
`
`
`
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`
`Ex. 2004, Vipperman Dep. Tr. at 41:6-25. Dr. Vipperman confirmed that the prior
`
`art does not disclose determination of a delay for one configuration that would enable
`
`beamforming for a plurality of configurations.
`
`8
`
`
`
`
`
`IPR2022-00004
`IPR2022-00004
`PATENT NO. RE47,049
`PATENT NO. RE47,049
`
`Q.
`
`And so looking at the prior art,
`
`Van Trees and Jeong, neither of those
`
`references disclose a determination of a
`
`delay for one geometry that enables
`
`beamforming for a different geometry.
`
`Right?
`
`MR. TUCKER: Objection to form.
`
`A.
`
`I think that's right.
`
`I mean,
`
`18
`
`11
`
`12
`
`13
`
`you could do this for any geometry,
`
`including the ones that -- the
`
`illustrative geometries that I pulled out
`
`of Van Trees in paragraphs 88 and 89 in my
`
`14
`
`declaration.
`
`Id., at 42:2-14.
`Id., at 42:2-14.
`
`
`
`9
`
`
`
`
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`
`Id., at 47:10-19. Thus, Dr. Vipperman acknowledges that his opinion is that a
`
`different delay would need to be determined for each configuration. This position is
`
`inconsistent with Dr. Vipperman’s declaration and the Petition, which do not
`
`identify a “plurality of configurations” each with their own respective delay
`
`determinations, as per the requirements set forth in Dr. Vipperman’s deposition
`
`testimony. See Pet. at 34-36; Ex. 1006, ¶¶ 88-89.
`
`V. THE LAW OF OBVIOUSNESS
`The question of obviousness is resolved on the basis of underlying factual
`
`determinations, including: (1) the scope and content of the prior art, (2) any
`
`10
`
`
`
`
`differences between the claimed subject matter and the prior art, (3) the level of skill
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`in the art, and (4) where in evidence, so called secondary considerations. Graham v.
`
`John Deere Co. of Kan. City, 383 U.S. 1, at 17–18 (1966). The question is not
`
`whether the differences themselves would have been obvious, but whether the
`
`claimed invention as a whole would have been obvious. Stratoflex, Inc. v. Aeroquip
`
`Corp., 713 F.2d 1530, 1537 (Fed. Cir. 1983).
`
`In arriving at an obviousness determination, the Board must sufficiently
`
`explain and support the conclusions that the prior art references disclose all the
`
`elements recited in the Challenged Claims and a relevant skilled artisan not only
`
`could have made but would have been motivated to combine all the prior art
`
`references in the way the patent claims and reasonably expected success. Pers. Web
`
`Techs., LLC v. Apple, Inc., 848 F.3d 987, 994 (Fed. Cir. 2017). That is, even if all
`
`the claim elements are found across a number of references, an obviousness
`
`determination must consider whether a person of ordinary skill in the art would have
`
`the motivation to combine those references. Intelligent Bio-Sys., Inc. v. Illumina
`
`Cambridge Ltd., 821 F.3d 1359, 1368 (Fed. Cir. 2016); Los Angeles Biomedical
`
`Research Inst. At Harbor-UCLA Med. Ctr. V. Eli Lilly & Co., 849 F.3d 1049, 1067
`
`(Fed. Cir. 2017) (vacating and remanding an obviousness determination, in part,
`
`because the Board did not make factual finding as to whether there was an apparent
`
`reason to combine all three prior art references to achieve the claimed invention and
`
`11
`
`
`
`
`whether a person of skill in the art would have had a reasonable expectation of
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`success from such a combination.) This combinability determination, as supported
`
`by an articulated motivation to combine, requires a plausible rationale as to why
`
`those prior art references would have worked together. Broadcom Corp. v. Emulex
`
`Corp., 732 F.3d 1325, 1335 (Fed. Cir. 2013).
`
`The Board has held that a failure to identify the differences between the
`
`claimed subject matter and the prior art is fatal to an obviousness challenge. See
`
`Apple, Inc. v. Contentguard Holdings, Inc., IPR2015-00355, Paper 9 at 9-10
`
`(P.T.A.B. June 26, 2015).
`
`VI. THE PROPOSED JEONG COMBINATIONS DO NOT
`DISCLOSE OR RENDER OBVIOUS THE DIGITAL SIGNAL
`PROCESSOR LIMITATIONS, INCLUDING THE THREE
`RECITED UNITS INTEGRATED IN A SINGLE DIGITAL
`SIGNAL PROCESSOR
`Claim 1 requires “wherein said sound source localization unit, said adaptive
`
`beamforming unit, and said noise reduction unit are integrated in a digital signal
`
`processor.” Ex. 1001, claim 1. There is no dispute that the claimed single digital
`
`signal processor must contain at least all three units as claimed. Ex. 2005, ¶ 29.
`
`First, Jeong fails to disclose or even suggest the claimed sound source
`
`localization unit, adaptive beamforming unit, and noise reduction unit integrated in
`
`a digital signal processor. Ex. 2005, McAlexander Decl., ¶ 30. Jeong does not recite
`
`the words “processor,” “microprocessor,” or “CPU” anywhere in its disclosure.
`
`12
`
`
`
`
`Accordingly, Petitioner’s expert states that Jeong discloses a “continuous process”
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`to conclude, without support, that the signal processing units in Jeong are integrated
`
`in a DSP. Ex. 1006 at ¶ 70 (“This is because Jeong’s apparatus obtains input mixed
`
`sound signals from the environment and outputs a target sound signal from which a
`
`noise is removed, which means that the signal processing between the input and the
`
`output must be continuous. Indeed, Jeong describes the signal processing between
`
`the sound source searching unit (223), the emphasized signal beam-former (221),
`
`the suppressed beam-former (222), and the signal extracting unit (230), as a
`
`continuous process.”); Ex. 2005, ¶ 31. However, that Jeong purportedly discloses a
`
`“continuous process” does not support his conclusion that Jeong discloses
`
`integration of all three units in a single digital signal processor. Id., ¶ 31.
`
`Further, Dr. Vipperman concludes that because a DSP is required to execute
`
`signal processing algorithms, “one skilled in the art would have understood that
`
`Jeong’s processing units are incorporated in a signal DSP.” Id. at ¶ 71. Again, Jeong
`
`does not disclose the integration of all three units in a single digital signal processor.
`
`Ex. 2005, ¶ 32. Further, there is no evidence of record to suggest that this critical
`
`feature of implementing or integrating the three claimed units in a single DSP is
`
`inherent or necessarily required by the mere presence of any feature(s) of Jeong. Id.,
`
`¶ 33. For example, as explained by Mr. McAlexander, a person of ordinary skill in
`
`the art would have recognized that the prior art systems at the time of the invention
`
`13
`
`
`
`
`used multiple separate circuits or processors to implement each unit. Ex. 2005, ¶ 33.
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`Recognizing that Jeong fails to disclose this limitation, Dr. Vipperman
`
`contends one skilled in the art would have found it obvious to integrate the signal
`
`processing units of Jeong in a DSP. Ex. 1006 at ¶ 72; Ex. 2005, ¶ 34. However, Dr.
`
`Vipperman cites to no disclosures of Jeong nor a motivation to combine to reach this
`
`conclusion. Ex. 2005, ¶ 34. Like IPR2020-00864, the primary reference does not
`
`show a single digital signal processor integrating all three recited units, and
`
`Petitioner does not make any attempt to explain how the units were integrated into a
`
`single DSP. Id.
`
`Alternatively, Petitioner relies on Briere to meet this limitation. See Ex. 1006
`
`at ¶ 73. Like Jeong, Briere does not disclose all three recited units integrated into a
`
`digital signal processor. Ex. 2005, ¶ 35. In addition, Briere does not cure the
`
`deficiencies of Jeong where Briere teaches away from using a DSP. Briere expressly
`
`states that “[i]t appears evident now that it will be impossible to build the original
`
`system on the DSP with the exact same parameters.” Ex. 1010 at 2; Ex. 2005, ¶ 35.
`
`Briere speculates as to what steps could “probably” take to build the system on to a
`
`DSP, which involves “reducing” the localization system. Ex. 1010 at 3; Ex. 2005, ¶
`
`35. Briere discloses that in 2006, it was undesirable to use a DSP in the manner set
`
`forth in the claimed invention and expressly taught away from use of DSP.
`
`Petitioner’s proposal relies on the addition of algorithms that it contends make up
`
`14
`
`
`
`
`the three cited units. Ex. 2005, ¶ 35. However, Petitioner’s proposal that a person
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`of ordinary skill in the art would have been capable of adding certain algorithms to
`
`a digital signal processor is insufficient to render obvious the claimed invention,
`
`particularly where the mere capability or possibility is not enough to establish prima
`
`facie obviousness. Id. Further, the reference’s admissions of impossibility and
`
`inoperability that it would have required “reducing” the recited units is both evidence
`
`of teaching away from the claimed invention and evidence against making the
`
`proposed modification. Id.
`
`Petitioner, nor its expert, Dr. Vipperman, does not address Briere’s admission
`
`of impossibility and interoperability which is fatal to Petitioner’s conclusory
`
`obviousness arguments. Accordingly, Petitioner’s contention that the three recited
`
`units integrated in a digital signal processor is disclosed by Jeong and Briere is
`
`incorrect.
`
`Second, a person of ordinary skill in the art would not be motivated to
`
`combine Jeong with Briere. Contrary to Petitioner’s assertion, (1) Briere teaches
`
`away from use of a digital signal processor, (2) any alleged prior art teaching of all
`
`claimed units on a single digital signal processor is not enabled by the prior art; and
`
`(3) the Briere reference is express evidence against any reason or motivation to
`
`combine because a person of ordinary skill in the art would have recognized that
`
`Briere’s digital signal processor was inoperative and that such a combination would
`
`15
`
`
`
`
`have been “impossible.” See Ex. 2005, ¶ 37; Ex. 1010. As confirmed by Petitioner’s
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`expert, Briere contemplates different aspects of the invention being carried out by
`
`different computing components and was unable to optimize the system for sound
`
`source localization, tracking, and separation on a single digital signal processor. Ex.
`
`2004, Vipperman Dep. Tr. at 64:8-66:18 (“You are right, they didn’t get it to
`
`work.”); Ex. 2005, ¶ 38. The solutions offered by Briere include “reducing sampling
`
`size, reducing resolution of the localization system, reducing maximum number of
`
`simultaneous sources.” Ex. 2004 at 66:19-67:11; Ex. 2005, ¶ 38. Briere also does
`
`not contemplate a different DSP. Ex. 2004 at 67:23-25; Ex. 2005, ¶ 38. Dr.
`
`Vipperman provides no evidence to the contrary. Ex. 2005, ¶ 38.
`
`The remainder of Petitioner’s reasons to combine Jeong and Briere fail for the
`
`same reasons and because they are based on mischaracterizations. As shown above,
`
`there
`
`is not one
`
`instance of “digital signal processor,” “processor,”
`
`“microprocessor,” or “CPU” in the Jeong reference. Ex. 2005, ¶ 39. Jeong makes no
`
`reference to sound source localization. Ex. 2005, ¶ 39. Petitioner offers no
`
`explanation for its motivation to combine Jeong, which does not disclose a DSP,
`
`with Briere, which teaches away from a single DSP. Id. Petitioner’s misleading
`
`characterizations are not supported in fact. Id. Accordingly, there is no motivation
`
`to combine Jeong and Briere to arrive at the claimed invention without
`
`impermissible hindsight, and there is no expectation of success. Id.
`
`16
`
`
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`
`VII. JEONG DOES NOT MEET THE “DETERMINING A DELAY”
`LIMITATIONS
`Claim 1 of the ’049 Patent recites “determining a delay between each of said
`
`sound sensors and an origin of said array of said sound sensors as a function of
`
`distance between each of said sound sensors and said origin, a predefined angle
`
`between each of said sound sensors and a reference axis, and an azimuth angle
`
`between said reference axis and said target sound signal, when said target sound
`
`source that emits said target sound signal is in a two dimensional plane, wherein said
`
`delay is represented in terms of number of samples, and wherein said determination
`
`of said delay enables beamforming for said array of sound sensors in a plurality of
`
`configurations.” Jeong fails to disclose or suggest “determining a delay.” Ex. 2005,
`
`¶ 41.
`
`Jeong discloses multiple delays and Dr. Vipperman’s mapping of this claim
`
`term is inconsistent. Id., ¶¶ 41-42. During his deposition, Dr. Vipperman stated that
`
`he has mapped both the “delaying unit” and the disclosed TDOA to this claim
`
`limitation. Ex. 2004, Vipperman Dep. Tr. at 72:10-18 (“Q. So it’s your opinion that
`
`the delay claimed in the ’049 and ’371 Patents is taught specifically by the TDOA
`
`method disclosed by Jeong? Is that fair? A. That’s correct.”); 75:18-25 (“Q. So if
`
`you look at paragraph 35—I am just trying to understand how this maps to the claims
`
`in your opinion. It says Figure 3B, the delaying unit 330 determines the delay term.
`
`17
`
`
`
`
`That is what you are saying is the determination of the delay in the challenged
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`claims. Right? A. That’s right.”); Ex. 2005, ¶ 42. Dr. Vipperman’s inconsistent
`
`positions with respect to Jeong and the delay limitation is fatal to the Petition. Jeong
`
`and Van Trees do not disclose or suggest each and every claim limitation related to
`
`the “determination of delay” and there is no motivation to combine Jeong with Van
`
`Trees to arrive at the claimed invention without impermissible hindsight, and there
`
`is no reasonable expectation of success. Ex. 2005, ¶ 43.
`
`The ”delaying unit” of Jeong is not the claimed invention, the TDOA
`
`disclosed in Jeong is not the same TDOA, and it is not used to arrive at the claimed
`
`invention. Dr. Vipperman and Petitioner attempt to import the “delaying unit” into
`
`the TDOA, but they are unrelated teachings. Id., ¶ 44.
`
`Dr. Vipperman also asserts that “the delaying unit” of Jeong to disclose the
`
`delay determination limitation. Id. at 75:18-25. Jeong discloses that the “delaying
`
`unit (430) delays the input signal X2(t) by a certain time via an adjustment of the
`
`adaptive delay term.” Ex. 1008 at ¶ 44. While Dr. Vipperman asserts that Jeong
`
`discloses an equation for determining a delay, he concedes that Jeong’s equation
`
`includes an “adjustment variable” which is “introduced to define the relationship
`
`between the sound pressure field and the directional control factors,” (76:1-77:3),
`
`but Jeong “stops short of telling us how to compute that.” Id. at 79:24-80:3.
`
`Petitioner and Dr. Vipperman admit that Jeong’s delay calculations do not
`
`18
`
`
`
`
`meet this limitation where they attempt to combine the disclosures of Van Trees with
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`Jeong. Pet. at 29-30. However, Petitioner offers no motivation to combine Jeong
`
`with Van Trees, where TDOA disclosed in Jeong and Van Trees are not the same.
`
`Ex. 2005, ¶ 45. As shown above, Jeong’s TDOA is a sound source location search
`
`method, and Van Trees discloses what Petitioner alleges is “calculations used to
`
`measure TDOA.” Pet. at 31; Ex. 2005, ¶ 46. Jeong’s equation fails to teach or suggest
`
`the determination of a delay in the Challenged Claims and Petitioner acknowledges
`
`as much by arguing Jeong discloses a “higher-level disclosure of calculating
`
`TDOA.” Pet. at 31; Ex. 2005, ¶ 46. However, merely identifying Van Trees’
`
`calculation and Jeong’s failure to specify a method by which to determine a delay is
`
`insufficient to provide a motivation to combine. Ex. 2005, ¶ 46.
`
`VIII. JEONG DOES NOT RENDER OBVIOUS THE “PLURALITY
`OF CONFIGURATIONS” LIMITATION
`Claim 1 of the ’049 Patent recites “determining a delay between each of said
`
`sound sensors and an origin of said array of said sound sensors as a function of
`
`distance between each of said sound sensors and said origin, a predefined angle
`
`between each of said sound sensors and a reference axis, and an azimuth angle
`
`between said reference axis and said target sound signal, when said target sound
`
`source that emits said target sound signal is in a two dimensional plane, wherein said
`
`delay is represented in terms of number of samples, and.” Jeong fails to disclose or
`
`19
`
`
`
`
`suggest “wherein said determination of said delay enables beamforming for said
`
`IPR2022-00004
`PATENT NO. RE47,049
`
`array of sound sensors in a plurality of configurations.”
`
`Petitioner fails to provide a motivation to combine Jeong’s sound source
`
`searching unit, which “determines a direction and spatial position of a target sound
`
`source,” with the “simple beamforming operation” disclosed in Van Trees to meet
`
`the “plurality of configurations” limitation. Pet. at 34; Ex. 2005, ¶ 49.
`
`Petitioner and Dr. Vipperman rely on Jeong in view of Van Trees. Pet. at 34
`
`(“Jeong in view of Van Trees discloses this limitation. As discussed above, Jeong
`
`discloses that the sound source searching unit (223) determines a direction and a
`
`spatial position of a target sound source, which is the process of beamforming. Van
`
`Trees also describes the process of determining time delay as an illustration of ‘a
`
`simple beamforming operation.’”) (internal citations omitted); Ex. 2005, ¶ 50.
`
`Petitioner concedes that Jeong discloses a linear array configuration and contends
`
`Van Trees discloses a plurality of array configurations. Ex. 2005, ¶ 50. However,
`
`Petitioner and its expert offer no explanation for how the determination of delay
`
`enables beamforming for a plurality of configurations nor the motivation to combine
`
`Jeong with Van Trees to meet this limitation. Id.
`
`Dr. Vipperman offered no explanation for how Jeong and Van Trees would
`
`enable beamforming for a plurality of configurations. Rather, during his deposition,
`
`Dr. Vipperman testified that if the configuration of a microphone array system were
`
`20
`
`
`
`
`changed, “[e]ach geometry will have its own specific delays that enable
`
`IPR20