`571-272-7822
`
`
`
`
`Paper 10
`Entered: December 28, 2023
`
`
`
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`____________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`____________
`
`META PLATFORMS, INC.,
`Petitioner,
`v.
`
`JAWBONE INNOVATIONS, LLC,
`Patent Owner.
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`
`
`Before GEORGIANNA W. BRADEN, NORMAN H. BEAMER,
`and JASON M. REPKO, Administrative Patent Judges.
`
`BEAMER, Administrative Patent Judge.
`
`DECISION
`Denying Institution of Inter Partes Review
`35 U.S.C. § 314
`Granting Unopposed Motion to Withdraw Motion for Joinder
`37 C.F.R. § 42.122
`
`
`
`
`
`
`
`
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`INTRODUCTION
`I.
`On June 27, 2023, Meta Platforms, Inc. (“Petitioner”) filed a Petition
`(“Pet.”) pursuant to 35 U.S.C. §§ 311–319 to institute an inter partes review
`of claims 1–20 of U.S. Patent No. 11,122,357 B2 (Ex. 1001, “the ’357
`patent”). Paper 1. On October 12, 2023, Jawbone Innovations, LLC
`(“Patent Owner”) filed a Preliminary Response (“Prelim. Resp.”). Paper 9.
`On June 27, 2023, Petitioner also filed a Motion for Joinder to And
`Consolidation With Related Inter Partes Review IPR2023-00251, which
`Patent Owner opposed on July 27, 2023. Papers 2, 7. However, on August
`8, 2023, that Proceeding was terminated due to settlement. IPR2023-00251,
`Paper 15. Accordingly, on September 8, 2023, Petitioner filed an
`Unopposed Motion To Withdraw Motion For Joinder. Paper 8. We grant
`that motion.
`The standard for instituting an inter partes review is set forth in
`35 U.S.C. § 314(a), which provides that an inter partes review may not be
`instituted unless the information presented in the Petition and any
`preliminary response shows that “there is a reasonable likelihood that the
`petitioner would prevail with respect to at least 1 of the claims challenged in
`the petition.” Applying that standard, we deny the Petition and do not
`institute an inter partes review.
`
`II. BACKGROUND
`The ’357 Patent
`A.
`The ’357 patent, titled “Forming Virtual Microphone Arrays Using
`Dual Omnidirectional Microphone Array (DOMA),” was filed on August 5,
`2013, issued on September 14, 2021, is a continuation of an application filed
`June 13, 2008, and lists several provisional applications, the earliest of
`2
`
`
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`which was filed on June 13, 2007. Ex. 1001, codes (54), (22), (45), (63),
`(60); Pet. 22.
`The ’357 patent is directed to:
`[a] dual omnidirectional microphone array noise suppression
`[system] used to form two distinct virtual directional
`microphones which are configured to have very similar noise
`responses and very dissimilar speech responses. The only null
`formed is one used to remove the speech of the user from [the
`second virtual microphone]. The two virtual microphones may
`be paired with an adaptive filter algorithm and VAD [Voice
`Activity Detector] algorithm to significantly reduce the noise
`without distorting the speech, significantly improving the SNR
`[signal-to-noise ratio] of the desired speech over conventional
`noise suppression systems.
`Ex. 1001, code (57). Figure 1 is reproduced below.
`
`
`Figure 1 is a block diagram of a two-microphone adaptive noise suppression
`system. Ex. 1001, 2:29–30. Microphones Mic 1 and Mic 2 receive acoustic
`information from speech signal source 101 and noise source 102, and the
`acoustic information received at each microphone is provided to Noise
`Removal system 105. Id. at 6:11–25. VAD 106 is a voice activity detector
`
`
`
`3
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`which generates a voicing information signal indicating when user speech is
`detected — for example, a skin surface microphone. Id. at 5:64–67, 6:5–6.
`Noise Removal component 105 generates the virtual microphones paired
`with the adaptive filter algorithm to generate cleaned speech 107. Id.
`at 5:16–21.
`The DOMA provides adaptive noise cancellation by filtering and
`summing the two microphone signals in the time domain. Ex. 1001,
`8:27–30. The adaptive filter generally uses the signal received from a
`microphone of the DOMA to remove noise from the speech received from
`the other microphone of the DOMA, relying on a slowly varying linear
`transfer function between the two microphones for sources of noise. Id. at
`8:30–35. Following processing of the two channels of the DOMA, an output
`signal is generated in which the noise content is attenuated with respect to
`the speech content. Id. at 8:35–38. According to the ’357 patent, the
`disclosed embodiments “result[] in excellent noise suppression performance
`and minimal speech removal and distortion.” Id. at 8:16–18.
`In one embodiment, virtual directional microphones are formed from
`the physical microphones as shown in Figure 4 reproduced below:
`
`
`
`4
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`
`Figure 4 is a block diagram for a DOMA depicting how two virtual
`directional microphones V1 and V2 are formed based on two physical
`omnidirectional microphones O1 and O2, where the physical microphone
`signals are coupled to processing component 402, in which delays z21, z11,
`z22, and z12, and gains A21, A11, A22, and A12 are applied to the signals, which
`are then summed via ∑1 and ∑2. Ex. 1001, 2:40–42, 8:62–9:21. As stated in
`the ’357 patent, “varying the magnitude and sign of the delays and gains of
`the processing paths leads to a wide variety of virtual microphones (VMs),
`also referred to herein as virtual directional microphones.” Id. at 9:22–25.
`In particular, for adaptive noise suppression, the delay and gain values
`are selected so that the noise responses of V1 and V2 are substantially
`similar, there is sufficient speech response for V1, and there is a relatively
`small speech response for V2, which ensures that the cleaned speech will
`have significantly higher signal-to-noise ratio than the original speech
`5
`
`
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`captured by microphone O1. Ex. 1001, 10:44–60. The ’357 patent discloses
`the results of a particular selection of delay and gain parameters that meet
`the criteria for effective noise suppression, in which arranging for the
`response of V2 to have a null at the speech location will cause V2 to exhibit
`minimal response to the speech. Id. at 11:50–52. The ’357 patent then
`explains that the speech null at zero degrees is not present for noise in the far
`field for the same microphone, with a noise source distance of
`approximately 1 meter. Id. at 11:54–57. The ’357 patent states this ensures
`that noise in front of the user will be detected so that it can be removed. Id.
`at 11:57–58. According to the ’357 patent, this differs from conventional
`systems that can have difficulty removing noise in the direction of the mouth
`of the user. Id. at 11:58–60.
`The plotted responses of the virtual microphones for an optimal
`selection of the delay and gain parameters are shown in a series of figures —
`first, in Figure 9 reproduced below.
`
`
`Figure 9 is a plot of linear response of virtual microphone V2 with delay=0.8
`to a 1 kHz speech source at a distance of 0.1 m. Ex. 1001, 11:40–42. The
`null in the linear response of virtual microphone V2 to speech is located at 0
`
`
`
`6
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`degrees, where the speech is typically expected to be located. Id. at 11:42–
`44.
`
`Next, the response of V2 to noise is shown in Figure 10 reproduced
`below.
`
`
`Figure 10 is a plot of a linear response of virtual microphone V2 with
`delay=0.8 to a 1 kHz noise source at a distance of 1.0 m. Id. at 11:44–47.
`The ’357 patent states that “[t]he linear response of V2 to noise is devoid of
`or includes no null, meaning all noise sources are detected.” Id. at 11:47–49.
`Furthermore, the response of V1 to speech is shown in Figure 11
`reproduced below.
`
`
`
`
`
`7
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`Figure 11 is a plot of linear response of virtual microphone V1 to a 1 kHz
`speech source at a distance of 0.1 m. Ex. 1001, 12:39–41. The ’357 patent
`states that “[t]here is no null and the response for speech is greater than that
`shown in FIG. 9.” Id. at 2:64–65.
`Finally, the response of V1 to noise is shown in Figure 12 reproduced
`below.
`
`
`Figure 12 is a plot of linear response of virtual microphone V1 to a 1 kHz
`noise source at a distance of 1.0 m. Ex. 1001, 12:44–46. The ’357 patent
`states that “[t]here is no null and the response is very similar to V2 shown in
`FIG. 10.” Id. at 3:1–2.
`
`Illustrative Claim
`B.
`Independent claim 1 is illustrative of the challenged subject matter
`and is reproduced below.
`1. A device, comprising:
`a first virtual microphone comprising a first
`combination of a first microphone signal and a
`second microphone signal, wherein the first
`microphone signal is generated by a first physical
`microphone and the second microphone signal is
`generated by a second physical microphone;
`
`
`
`8
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`a second virtual microphone comprising a second
`combination of the first microphone signal and
`the second microphone signal, wherein the
`second combination is different from the first
`combination, wherein the first virtual
`microphone and the second virtual microphone
`are distinct virtual directional microphones with
`substantially similar responses to noise and
`substantially dissimilar responses to speech; and
`a signal processor coupled with the first and second
`microphone signals and operative to combine the
`first and second microphone signals by filtering
`and summing in the time domain, to apply a
`varying linear transfer function between the first
`and second microphone signals, and to generate
`an output signal having noise content that is
`attenuated with respect to speech content.
`Ex. 1001, 34:55–35:10.
`We note that Figures 9–12 reproduced above exemplify the claim 1
`requirement that the virtual microphones have “substantially similar
`responses to noise and substantially dissimilar responses to speech.”
`
`
`
`9
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`Asserted Challenges to Patentability and Evidence of Record
`C.
`Petitioner challenges the patentability of claims 1–20 of the ’357
`patent based on the following combination of references:
`Claims Challenged
`35 U.S.C. §
`Reference(s)/Basis
`1–20
`1031
`Brandstein,2 Gannot,3
`Brandstein, Gannot, Griffiths-
`4, 13, 14, 18, 19
`103
`Jim 4
`4, 5, 6, 13, 14, 18, 19
`103
`Brandstein, Gannot, McCowan5
`Pet. 22. In support of its patentability challenge, Petitioner relies on, inter
`alia, the Declaration of Richard M. Stern, Ph.D. that was filed in IPR2023-
`00251 and refiled in this Proceeding. Ex. 1002 (“Stern Decl.”).
`
`
`1 The Leahy-Smith America Invents Act, Pub. L. No. 112-29, 125 Stat. 284
`(2011) (“AIA”), included revisions to 35 U.S.C. § 103 that became effective
`as of March 16, 2013. Although the application for the ’357 patent was filed
`after March 16, 2013, it includes a priority claim to an application filed
`before this date. Ex. 1001, codes (22), (60), (63). Accordingly, for purposes
`of institution, we apply the pre-AIA version of 35 U.S.C. § 103.
`2 Brandstein, et al. (ed.), Microphone Arrays: Signal Processing Techniques
`And Applications (excerpts) (2001) (“Brandstein,” Ex. 1003).
`3 Gannot et al., Signal Enhancement Using Beamforming and
`Nonstationarity with Applications to Speech, Vol. 49, No. 8, IEEE
`Transactions On Signal Processing, 1614 (Aug. 2001) (“Gannot,” Ex. 1004).
`4 Griffiths & Jim, An Alternative Approach to Linearly Constrained
`Adaptive Beamforming, Vol. AP-30, No. 1, IEEE Transactions On Antennas
`And Propagation, 27 (Jan. 1982) (“Griffiths-Jim,” Ex. 1005).
`5 Iain A. McCowan, et al., Near-Field Adaptive Beamformer for Robust
`Speech Recognition, Digital Signal Processing, Vol. 12, Issue 1 (2002), 87–
`106 (“McCowan,” Ex. 1006).
`
`
`
`10
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`Real Parties in Interest
`D.
`The parties identify themselves as real parties in interest. Pet. 80;
`Paper 6, 2.
`
`Related Proceedings
`E.
`The parties identify the following related proceedings: Jawbone
`Innovations, LLC v. Samsung Electronics Co., Ltd. and Samsung Electronics
`America, Inc., Case No. 2:21-cv-00186-JRG (E.D. Tex.); Jawbone
`Innovations, LLC v. Amazon.com, Inc., Case No. 5:22-cv-06727-TLT (N.D.
`Cal.); Jawbone Innovations, LLC v. Amazon.com, Inc., Case No. 2:21-cv-
`00435 (E.D. Tex.); Jawbone Innovations, LLC v. Apple Inc., Case No. 6:21-
`cv-00984-ADA (W.D. Tex.); Jawbone Innovations, LLC v. Google LLC,
`Case No. 6:21-cv-00985-ADA (W.D. Tex.); Jawbone Innovations, LLC v.
`Google LLC, Case No. 3:23-cv-00466 (N.D. Cal.); Jawbone Innovations,
`LLC v. Guangdong OPPO Mobile Telecommunications Corp., Ltd., d/b/a
`OPPO, Case No. 2:23-cv-00079 (E.D. Tex.); Jawbone Innovations, LLC v.
`HTC Corporation, Case No. 2:23-cv-00077 (E.D. Tex.); Jawbone
`Innovations, LLC v. LG Electronics, Inc., Case No. 2:23-cv-00078 (E.D.
`Tex.); Jawbone Innovations, LLC v. Meta Platforms, Inc., Case No. 6:23-cv-
`00158 (W.D. Tex.); Jawbone Innovations, LLC v. Panasonic Holdings
`Corporation, Case No. 2:23-cv-00081 (E.D. Tex.); Jawbone Innovations,
`LLC v. Sony Electronics Inc., Case No. 2:23-cv-01161-MEF-LDW (D.N.J.);
`Jawbone Innovations, LLC v. ZTE Corporation, Case No. 2:23-cv-00082
`(E.D. Tex.); Google LLC v. Jawbone Innovations, LLC, IPR2022-01124;
`Samsung Electronics Co., Ltd., et al. v. Jawbone Innovations, LLC,
`IPR2022-01321; Amazon.Com, Inc. v. Jawbone Innovations, LLC, IPR2023-
`00251. Pet. 80–81; Paper 6, 2–3.
`
`
`
`11
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`III. ANALYSIS OF PETITIONER’S CHALLENGE
`Legal Standards
`A.
`To prevail in its challenge, Petitioner must demonstrate by a
`preponderance of the evidence that the claims are unpatentable. 35 U.S.C.
`§ 316(e); 37 C.F.R. § 316(e). “In an IPR, the petitioner has the burden from
`the onset to show with particularity why the patent it challenges is
`unpatentable.” Harmonic Inc. v. Avid Tech., Inc., 815 F.3d 1356, 1367 (Fed.
`Cir. 2016) (citing 35 U.S.C. § 312(a)(3) (2012) (requiring inter partes
`review petitions to identify “with particularity . . . the evidence that supports
`the grounds for the challenge to each claim”)). This burden of persuasion
`never shifts to the patent owner. See Dynamic Drinkware, LLC v. Nat’l
`Graphics, Inc., 800 F.3d 1375, 1378 (Fed. Cir. 2015) (discussing the burden
`of proof in inter partes review).
`A patent claim is unpatentable under 35 U.S.C. § 103 if the
`differences between the claimed subject matter and the prior art are such that
`the subject matter, as a whole, would have been obvious at the time the
`invention was made to a person having ordinary skill in the art to which said
`subject matter pertains. KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 406
`(2007). 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 differences between the claimed subject matter and the prior art;
`(3) the level of ordinary skill in the art; and (4) when in evidence, objective
`indicia of non-obviousness (also called secondary considerations), such as
`commercial success, long-felt but unsolved needs, and failure of others.
`Graham v. John Deere Co., 383 U.S. 1, 17–18 (1966). We analyze grounds
`based on obviousness in accordance with the above-stated principles.
`12
`
`
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`Level of Ordinary Skill in the Art
`B.
`In determining whether an invention would have been obvious at the
`time it was made, 35 U.S.C. § 103(a) requires us to resolve the level of
`ordinary skill in the pertinent art at the time of the invention. Graham, 383
`U.S. at 17. The person of ordinary skill in the art is a hypothetical person
`who is presumed to have known the relevant art at the time of the invention.
`In re GPAC Inc., 57 F.3d 1573, 1579 (Fed. Cir. 1995). Factors that may be
`considered in determining the level of ordinary skill in the art include, but
`are not limited to, the types of problems encountered in the art, the
`sophistication of the technology, and educational level of active workers in
`the field. Id. In a given case, one or more factors may predominate. Id.
`Petitioner asserts that a person of ordinary skill in the art at the time of
`the alleged invention of the ’357 patent:
`[W]ould have had a minimum of a bachelor’s degree in
`computer engineering, computer science, electrical engineering,
`mechanical engineering, or a similar field, and approximately
`three years of industry or academic experience in a field related
`to acoustics, speech recognition, speech detection, or signal
`processing. . . . Work experience could substitute for formal
`education, and additional formal education could substitute for
`work experience.
`Pet. 24 (citing Stern Decl. ¶¶ 27–28). Patent Owner utilizes Petitioner’s
`proposed level of skill in the art “[f]or the purposes of this Preliminary
`Response only.” Prelim. Resp. 7.
`Petitioner’s proposal is consistent with the level of ordinary skill in
`the art reflected by the asserted prior art. See Okajima v. Bourdeau, 261
`F.3d 1350, 1355 (Fed. Cir. 2001); GPAC, 57 F.3d at 1579. On this record,
`
`
`
`13
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`the level of ordinary skill is neither in dispute nor dispositive of any
`challenge. For purposes of this Decision, we apply Petitioner’s articulation.
`
`C. Claim Construction
`The Petition was accorded a filing date of June 27, 2023. Paper 4, 1.
`In an inter partes review for a petition filed on or after November 13, 2018, a
`claim “shall be construed using the same claim construction standard that
`would be used to construe the claim in a civil action under 35 U.S.C. 282(b).”
`37 C.F.R. § 42.100(b). We apply the claim construction standard from
`Phillips v. AWH Corp., 415 F.3d 1303, 1312–13 (Fed. Cir. 2005) (en banc).
`Claim terms need only be construed to the extent necessary to resolve
`the controversy. Nidec Motor Corp. v. Zhongshan Broad Ocean Motor Co.
`Matal, 868 F.3d 1013, 1017 (Fed. Cir. 2017).
`The parties submit that no claim construction is necessary.
`Pet. 24–25; Prelim. Resp. 6. Therefore, we do not construe the claims.
`
`D.
`
`Alleged Obviousness of Independent Claims 1 and 15 over Brandstein
`and Gannot
`Petitioner challenges claims 1 and 15 as obvious over the combination
`of Brandstein and Gannot. Pet. 25–37, 61–65.
`
`1. Brandstein (Ex. 1003)
`Brandstein is a textbook titled “Microphone Arrays: Signal Processing
`Techniques And Applications,” copyrighted 2001. Ex. 1003. Petitioner
`provides evidence that Brandstein is prior art under pre-AIA §102(b).
`Pet. 23 (citing Ex. 1013 ¶¶ 19–30; Stern Decl. ¶ 44). Patent Owner raises no
`issue at this time as to the prior art status of this reference, and we treat it as
`prior art for purposes of this Decision.
`14
`
`
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`Brandstein discloses techniques for beamforming in microphone
`arrays with the “goal of speech enhancement . . . to remove undesirable
`signals such as noise and reverberation.” Ex. 1003, 87. Figure 5.1 is block
`diagram reproduced below.
`
`
`
`As indicated in the caption, Figure 5.1 is a “Griffiths-Jim beamformer,”
`which is a reference to a January 1982 paper by Griffiths and Jim titled “An
`Alternative Approach to Linear Constrained Adaptive Beamforming.” Ex.
`1003, 108 (citation 27); Ex. 1005, 27; Stern Decl. ¶ 46. The upper portion of
`the figure depicts a fixed beamformer that filters and sums inputs from
`microphone array x0(k)–xN-1(k) to generate an output representing a speech
`source (indicated as an arrow directed to the array at angle theta (θ)).
`Ex. 1003, 88; Stern Decl. ¶ 48. The lower portion of the figure depicts a
`blocking matrix and multiple input canceller that form a signal with a null in
`
`
`
`15
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`the direction of the speech so that speech is suppressed and noise signals are
`passed through. Ex. 1003, 88; Stern Decl. ¶ 49. The output of the lower
`portion is subtracted for that of the upper portion to produce y(k), in which
`speech “is enhanced and undesirable signals such as ambient noise and
`interferences are suppressed.” Ex. 1003, 88–89; Stern Decl. ¶ 50.
`
`2. Gannot (Ex. 1004)
`Gannot is an article titled “Signal Enhancement Using Beamforming
`and Nonstationarity with Applications to Speech,” included in the August
`2001 issue of IEEE Transactions On NAL Processing. Ex. 1004, 1614.
`Petitioner provides evidence that Gannot is prior art under pre-AIA §102(b).
`Pet. 23 (citing Ex. 1013 ¶¶ 31–42). Patent Owner raises no issue at this time
`as to the prior art status of this reference, and we treat it as prior art for
`purposes of this Decision.
`Gannot discloses sensor (e.g., microphone) arrays “where arbitrary
`transfer functions (TFs) relate the source signal and the sensors.” Ex. 1004,
`1614. Like Brandstein, Gannot refers to the Griffiths-Jim beamformer, but
`modifies the approach to account for varying transfer functions. Id.; Stern
`Decl. ¶ 56. Figure 3 is a circuit diagram reproduced below.
`
`
`
`16
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`
`Figure 3 depicts a beamformer similar to that of Figure 5.1 of Brandstein,
`but the blocking matrix (indicated as H †) uses estimates of time varying
`transfer function ratios of different sensors. Ex. 1004, 1615, 1618; Stern
`Decl. ¶¶ 57–61.
`
`3. Petitioner’s Challenge To Independent Claims 1 and 15
`For the “device” preamble of independent claim 1, Petitioner
`generally relies on Brandstein’s disclosure of a device with a microphone
`array and signal processing, including for use in a hearing aid or telephone.
`Pet. 25–26 (citing Ex. 1003, Figs. 5.1, 5.16, 87, 89, 104–105; Stern Decl.
`¶ 85). 6
`For the “first virtual microphone” requirement of claim 1, Petitioner
`relies on the top portion of Figure 5.1 of Brandstein, as shown in the
`annotated version of that figure reproduced below.
`
`
`6 Based on the present record, we make no determination that the preamble
`of claim 1 (or of claim 15) is limiting.
`17
`
`
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`
`Pet. 26–27 (citing Ex. 1003, Fig. 5.1, 89; Stern Decl. ¶ 89). Petitioner
`associates x0(k) and x1(k) as the first and second microphones, with the first
`virtual microphone as the output of the filtering and summing of the
`inputs. Id.
`For the “second virtual microphone” requirement of claim 1,
`Petitioner relies on the bottom portion of Figure 5.1 of Brandstein, as shown
`in the annotated version of that figure reproduced below.
`
`
`
`18
`
`
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`Pet. 27–28 (citing Ex. 1003, Fig. 5.1, 89; Stern Decl. ¶ 91). Regarding the
`requirement of “distinct virtual directional microphones with substantially
`similar responses to noise and substantially dissimilar responses to speech,”
`Petitioner relies on Figure 5.2 of Brandstein, an annotated version of which
`is reproduced below.
`
`
`
`Pet. 28–30 (citing Ex. 1003, Fig. 5.2, 87–90, 101; Stern Decl. ¶¶ 94–97).
`Petitioner’s declarant explains that the zero gain indicated for the “Target”
`(i.e., speech) compared to the “deep null” for “Interference,” discloses that
`the two virtual microphones are distinct, that their responses to noise are
`substantially similar (thus cancelling each other out as to noise), and that
`their responses to speech are substantially dissimilar (because no significant
`speech signal from the second virtual microphone is subtracted from the first
`virtual microphone signal). Stern Decl. ¶¶ 96–97.
`For the “signal processor” requirement of claim 1, Petitioner relies on
`Figure 5.1 of Brandstein as showing the microphone signals being combined
`by filtering and summing in the time domain, as shown, for example, by the
`components labeled “Fixed Filter” in the upper portion of the figure, and the
`
`
`
`19
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`component “⊕,” indicating summation. Pet. 32 (citing Stern Decl. ¶ 100).
`
`Petitioner’s declarant explains that the fact that the microphone signals are
`functions of discrete time “k” indicates that the operation is in the time
`domain. Id.
`For the requirement that the signal processor is operative to “apply a
`varying linear transfer function between the first and second microphone
`signals,” Petitioner relies on the disclosure in Gannot of a modified version
`of the Griffiths-Jim beamformer in which, in order to account for arbitrary
`acoustic paths, the blocking matrix uses ratios of varying transfer functions
`for the microphones. Pet. 33–35 (citing Ex. 1004, Figs. 3, 4, 1615–1620;
`Stern Decl. ¶¶ 105–106). Petitioner’s declarant testifies that one of ordinary
`skill would have been motivated to combine the teachings of Brandstein and
`Gannot because Gannot proposed a modification of the original Griffiths-
`Jim beamformer which is also the basis for the pertinent disclosure in
`Brandstein, which modification would have allowed for a more robust
`accommodation of situations where the microphone transfer functions were
`varying. Stern Decl. ¶¶ 106, 110.
`Finally, for the requirement that the signal processor is operative to
`“generate an output signal having noise content that is attenuated with
`respect to speech content,” Petitioner relies on the disclosure in Brandstein
`
`that for the output 𝑦𝑦(𝑘𝑘), “the target signal is enhanced and undesirable
`
`signals such as ambient noise and interferences are suppressed.” Pet. 36–37
`(citing Ex. 1003, 88–89; Stern Decl. ¶ 109).
`Independent claim 15 has the same requirements as claim 1, and in
`addition requires, “a virtual microphone array including the first and second
`virtual microphones and having a single null oriented in a direction toward a
`
`
`
`20
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`source of speech.” Ex. 1001, 36:8–32. For this additional requirement,
`Petitioner relies on Figure 5.3 of Brandstein, an annotated version of which
`is reproduced below.
`
`
`
`Pet. 62–63 (citing Ex. 1003, Fig. 5.3, 90; Stern Decl. ¶ 161). Figure 5.3
`shows the single null in the Blocking Matrix oriented towards the speech
`source. Id.
`
`4. Patent Owner’s Response To Petitioner’s Challenge To Independent
`Claims 1 and 15
`Patent Owner argues that Brandstein and Gannot do not render
`claims 1 or 15 obvious because the combination does not teach or suggest
`the independent claim 1 and claim 15 requirement, “wherein the first virtual
`microphone and the second virtual microphone are distinct virtual
`directional microphones with substantially similar responses to noise.”
`Prelim. Resp. 8–14. Patent Owner notes that, for this claim requirement,
`Petitioner and its declarant, Dr. Stern, rely solely on Figure 5.2 of
`Brandstein, which depicts an example directivity pattern of a Griffiths-Jim
`
`
`
`21
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`beamformer, and in particular shows a large negative gain in the direction of
`arrival of interference. Prelim. Resp. 12–13 (citing Pet. 29–30; Stern Decl.
`¶¶ 95, 97). Petitioner and its declarant deduce that the large negative gain in
`the interference direction shown on Figure 5.2 “reflects that subtracting the
`two virtual microphones’ outputs cancels the interference, which confirms
`that the virtual microphones’ responses to the interference are essentially the
`same.” Pet. 30 (citing Stern Decl. ¶ 97).
`However, argues Patent Owner, the assumption that Figure 5.2 shows
`that the virtual microphones have similar responses to noise is unwarranted,
`because Figure 5.2 shows the overall directivity of an exemplary Griffiths-
`Jim beamformer, not the individual directivity of either virtual microphone,
`thus reflecting the overall result of the system in attenuating noise, rather
`than disclosing anything about the comparative responses to noise of the
`virtual microphones. Prelim. Resp. 12–13.
`In addition, Patent Owner cites Brandstein’s description of Figure 5.1,
`which is the structure that Figure 5.2 relates to, and which states:
`Figure 5.1 depicts the structure of the GJBF [Griffiths-
`Jim beamformer]. It comprises a fixed beamformer (FBF), a
`multiple-input canceler (MC), and a blocking matrix (BM).
`The FBF is designed to form a beam in the look direction so
`that the target signal is passed and all other signals are
`attenuated. On the contrary, the BM forms a null in the look
`direction so that the target signal is suppressed and all other
`signals are passed through.
`Ex. 1003, 88 (cited at Prelim. Resp. 9–10). Patent Owner, relying on its
`declarant, argues that this excerpt demonstrates that the virtual microphones
`of Brandstein do not have “substantially similar responses to noise,” because
`the fixed beamformer (which Petitioner associates with the first virtual
`
`
`
`22
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`microphone of the claims) only passes the target signal (i.e., voice) and
`attenuates all other signals (i.e., noise), whereas the blocking
`matrix/multiple-input canceler combination (which Petitioner associates
`with the second virtual microphone) suppresses the target signal and passes
`through all other signals. Prelim. Resp. 10–11 (citing Ex. 2002 ¶¶ 42–43
`(Declaration Of Akbar M. Sayeed, Ph.D.)).7
`Patent Owner also relies on Figure 5.3 of Brandstein, reproduced
`below, with Patent Owner’s annotations.
`
`
`
`The upper portion of Figure 5.3 depicts the directivity of the fixed
`beamformer of Figure 5.1, and the upper portion depicts the directivity of
`the blocking matrix. Prelim. Resp. 13 (citing Ex. 1003, Fig. 5.3, 90;
`Ex. 2002 ¶ 45). Patent Owner’s declarant explains that shaded portions
`show that the noise responses of the two virtual microphones are not
`substantially similar. Ex. 2002 ¶ 45.
`
`
`7 The Preliminary Response erroneously cites the Sayeed declaration as
`Exhibit 2004. We understand that Exhibit 2002 is the intended citation.
`23
`
`
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`As a separate issue, Patent Owner points out that the Brandstein
`reference is a collection of different papers by different authors, each
`reproduced as a chapter, and argues that Petitioner does not provide any
`reason why one of ordinary skill would have combined the teachings from
`the different papers in the manner that Petitioner relies on. Prelim.
`Resp. 14–18.
`
`5. Analysis of Petitioner’s Challenge To Independent Claims 1 and 15
`We agree with Patent Owner that the record does not support
`Petitioner’s assertion that Brandstein teaches or suggests the independent
`claim requirement, “wherein the first virtual microphone and the second
`virtual microphone are distinct virtual directional microphones with
`substantially similar responses to noise.” Contrary to Petitioner’s
`interpretation of Brandstein Figure 5.2, that figure does not lead to an
`inference that the individual microphone responses have similar responses to
`noise. Prelim. Resp. 12–13. In effect, Petitioner is arguing that knowing
`that the sum of two numbers is equal to “2” necessarily means that each
`number is “1”, as opposed, for example, to one number being “2” and the
`other “0”.
`Moreover, as discussed above, Brandstein’s description of Figure 5.1
`demonstrates that the virtual microphones of Brandstein do not have
`“substantially similar responses to noise.” Prelim. Resp. 9–11. Likewise,
`Figure 5. 3 of Brandstein also refutes Petitioner’s assertion the virtual
`microphone noise responses are substantially similar. Prelim. Resp. 13. 8
`
`
`8 We need not address Patent Owner’s argument that Petitioner’s reliance on
`different chapters of Brandstein is unjustified.
`24
`
`
`
`
`
`IPR2023-01130
`Patent 11,122,357 B2
`
`
`Accordingly, we determine that the Petition does not demonstrate a
`reasonable likelihood that Petitioner would prevail in establishing that
`independent claims 1 and 15 are unpatentable under 35 U.S.C. § 103(a) over
`the combination of Brandstein and Gannot.
`
`E. Alleged Obviousness of Dependent Claims 2–14 and 16–20
`Petitioner also challenges dependent claims 2–14 and 16–20 as
`obvious over the combination of Brandstein and Gannot. Pet. 37–61, 65–67.
`In addition, Petitioner challenges dependent claims 4, 13, 14, 18, and 19 as
`obvious over the combination of Brandstein, Gannot, and Griffiths-Jim, and
`also challenges dependent claims 4, 5, 6, 13, 14, 18, and 19 as obvious over
`the combination of Brandstein, Gannot, and McCowan. Pet. 67–72. Neither
`Griffiths-Jim nor McCowan are relied on for the above-discussed
`independent claim limitat