`Petitioner,
`v.
`PARUS HOLDINGS, INC.,
`Patent Owner
`
`IPR2020-00686; Patent 7,076,431 B2
`IPR2020-00687; Patent 9,451,084 B2
`Paper 25 – Order Setting Oral Argument
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`1
`
`
`
`Grounds
`
`IPR2020-00686
`Ground Reference
`Ladd, Kurosawa, and Goedken
`I
`Ladd, Kurosawa, Goedken, and Madnick
`II
`Ladd, Kurosawa, Goedken, and Houser
`III
`Ladd, Kurosawa, Goedken, and Rutledge
`IV
`
`Basis Challenged Claims
`§103 1-6, 9-10, 13-14, 18, 20-21, and 25
`§103 7, 19, and 26-30
`§103 5 and 6
`§103 9 and 25
`
`IPR2020-00687
`Ground Reference
`Ladd, Kurosawa, and Goedken
`I
`Ladd, Kurosawa, Goedken, and Madnick
`II
`Ladd, Kurosawa, Goedken, and Houser
`III
`Ladd, Kurosawa, Goedken, and Rutledge
`IV
`Ladd, Kurosawa, Goedken, Rutledge, and Madnick
`V
`Ladd, Kurosawa, Goedken, Rutledge, and Houser
`VI
`
`Basis Challenged Claims
`§103
`1-6, 10, and 14
`§103
`7
`§103
`5 and 6
`§103
`1-6, 10, and 14
`§103
`7
`§103
`5 and 6
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`2
`
`
`
`“Speaker Independent Speech Recognition Device”
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`3
`
`
`
`Speaker Independent – All Grounds Fail
`
` All Grounds Fail Because The Petition Does Not Identify Any Reference That
`Teaches The “at least one speaker-independent speech recognition device,
`said speaker-independent speech recognition device operatively connected to
`said computer and to said voice enabled device” Limitation Found In Every
`Challenged Claim
`
`of Ladd, Kurosawa, and Goedken or by any other combination
`
`‒ The Petition does not establish that Claim 1(c) is taught by the combination
`‒ As mapped by the Petition, Claim 1(c) reads as follows:
`‒ At least one speaker-independent speech recognition device, said speaker-
`
`independent speech recognition device operatively connected to said
`computer and to said voice enabled device
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`4
`
`
`
`Every Ground Of The Petition Relies On Ladd To Teach The Speaker-Independent
`Speech Recognition Device (Petition, 22-25)
`
` Ladd’s disclosure of speaker-independent speech recognition is different and
`significantly inferior to the speaker-independent speech recognition device
`disclosed by the ’431 and ’084 Patents, therefore this claim limitation is not
`disclosed or taught by Ladd.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`5
`
`
`
`Claim Construction – “speaker-independent speech recognition device”
`
` “speaker-independent speech recognition device” should be construed as a “speech recognition
`device that recognizes spoken words without using predefined voice patterns”
` The ’431 and ’084 patents disclaimed the use of voice patterns to recognize a spoken command.
`
`Ex. 1001 (Kurganov)
`’431 Patent at 4:34-43
` The Western District of Texas in a parallel district court litigation Parus v. Apple, 6:19-00432-ADA,
`agreed with this understanding of the ’431 and ’084 Patents when it construed “speaker-independent
`speech recognition device” to mean “speech recognition device that recognizes spoken words
`without adapting to individual speakers or using predefined voice patterns.”
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`6
`
`
`
`Ladd Does Not Disclose Speaker-Independent In The Context Of The Application
`Described in the ’431 and ’084 Patents
`
` Ladd is an application where specific pre-defined voice patterns are
`recognized as disclaimed by the ’431 and ’084 Patents.
` Ladd does not recognize phonemes, and is not speaker-independent as
`specifically defined by the ’431 and ’084 Patents.
`
`Ex. 1004 (Ladd)
`’336 Patent at 9:36-39
`
` Ladd makes clear that its ASR recognizes a predefined speech pattern.
` Ladd does not describe identifying discrete segments in speech or voice
`patterns but identifies a selected speech pattern of the speech.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`7
`
`
`
`Ladd Does Not Disclose Speaker-Independent In The Context Of The Application
`Described in the ’431 and ’084 Patents
`
` Ladd makes clear that the audio input is compared to the vocabulary.
`
`Ex. 1004 (Ladd)
`’336 Patent at 10:3-20
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`8
`
`
`
`Ladd Does Not Disclose Speaker-Independent In The Context Of The Application
`Described in the ’431 and ’084 Patents
`
` If the ASR unit 254 could translate speech to text, as asserted by Apple,
`there is no need for a STT unit 256.
`
`Ex. 1004 (Ladd) ’336 Patent at 9:28-30
`
`Ex. 1004 (Ladd) ’336 Patent at 9:45-47
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`9
`
`Ex. 1004 (Ladd) ’336 Patent at Fig. 3
`
`
`
`“Sequential Access”
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`10
`
`
`
`None Of The Cited References Disclose Sequential Access
`
` In every Ground, the Petition relies on Goedken to provide this limitation, which is exemplary in
`the ’431 and mapped by Apple as follows:
`
` Goedken does not disclose sequentially accessing pre-selected web sites; rather, the Goedken
`disclosure relied on an identified by Petitioner discloses the accessing of internal database files,
`not web sites.
`
` The Petition expressly admits this failure conceding that Goedken is “searching files rather than
`websites.” (Petition, 46).
`
`Ex. 1001 (Kurganov)
`’431 Patent at 20:24-34
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`11
`
`
`
`Sequential Access Of Websites Is Very Different
`Than Sequential Access Of A Database
`
` A database is controlled by a system administrator. The administrator can
`update the data in the database whenever they please.
` In contrast, when one accesses a website one discovers information that is
`there at that moment.
` Information on websites can be dynamic and change often, and is out of the
`control of the user.
` Sequentially accessing websites where one cannot control the data that may
`be there at that moment is very different than accessing files in a database
`that are in a controlled environment under the control of the system
`administrator.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`12
`
`
`
`“No Motivation to Combine”
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`13
`
`
`
`There Is No Motivation To Combine Ladd With Kurosawa
`Or Ladd With Kurosawa And Goedken
`
` The Petition relies on impermissible hindsight in order to combine the references and the
`Federal Circuit has been clear that this is not proper:
` One cannot “stitch together an obviousness finding from discrete portions of prior art
`references without considering the references as a whole.”
`In re Enhanced Security Research, LLC, 739 F.3d 1347, 1355 (Fed. Cir. 2014).
`
` The “[d]etermination of obviousness cannot be based on the hindsight combination of
`components selectively culled from the prior art to fit the parameters of the patented
`invention.” ATD Corp. v. Lydall, Inc., 159 F.3d 534, 546 (Fed. Cir. 1998).
` “One cannot use hindsight reconstruction to pick and choose among isolated disclosures in the
`prior art to deprecate the claimed invention.” In re Fine, 837 F.2d 1071, 1075 (Fed. Cir. 1988).
` “It is impermissible, however, simply to engage in a hindsight reconstruction of the claimed
`invention, using the applicant’s structure as a template and selecting elements from the
`references to fill the gaps.” In re Gorman, 983 F.2d 982, 987 (Fed. Cir. 1991).
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`14
`
`
`
`There Is No Motivation To Combine Ladd With Kurosawa
`
` Ladd is an IVR system that has to mitigate latencies to provide near real-time communications.
` Kurosawa teaches a system that accesses data from a plurality of URLs, creates a search report with all
`of the data from a plurality of URLs, and returns the search report with all of the data from the plurality
`of URLs to the user.
`
`Ex. 1005 (Kurosawa Certified
`Translation) at Abstract
` Accessing a plurality of URLs instead of the single URL that Ladd accesses would create undue latencies
`in Ladd’s system and make it slower.
` A POSITA would not combine Kurosawa with Ladd to make Ladd slower.
` This is impermissible hindsight.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`15
`
`
`
`There Is No Motivation To Combine Ladd With Kurosawa And Goedken
`
` The Petition relies on Bennett as its motivation to combine Goedken with Ladd and
`Kurosawa indicating that real-time responses to spoken questions should be within
`3-5 seconds to provide the appearance of a real-time discourse. (Petition, 45).
` Combining Ladd with Kurosawa, where multiple URLs are accessed before
`returning a search report to the user, runs counter to the motivation in
`Bennett, because accessing multiple URLs before returning the search report
`would take much longer than 3-5 seconds and would not provide the
`appearance of real-time disclosure.
` The Petition suggests that it would be obvious to slow down Ladd with Kurosawa,
`and then speed it back up with Goedken.
` This is impermissible hindsight reconstruction.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`16
`
`
`
`Goedken Teaches Away From Kurosawa
`
` The explicit teachings of Goedken teach away from Kurosawa because Goedken teaches away from
`using web search engines, and Kurosawa is a specialized web search engine.
` “while Internet searching is a powerful resource, it is often inferior to having the capability of posing
`a direct question to a human expert who can provide a direct answer.”
`
`Ex. 1006 (‘Goedken)
`’423 Patent at 1:51-54
` “[s]earch engines results can be over-inclusive, overwhelming, and may not contain the
`appropriate information.”
`
`Ex. 1006 (‘Goedken)
`’423 Patent at 5:22-25
` Apple uses the basic structure of Ladd, an IVR system, as a template and selects elements from
`Kurosawa and Goedken to fill in the gaps.
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`17
`
`
`
`End
`
`DEMONSTRATIVE EXHIBIT – NOT EVIDENCE
`
`Parus IPR-
`
`18
`
`