`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
` UNITED STATES PATENT AND TRADEMARK OFFICE
`
`
`
`
`
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`
`
`
`
`
`APPLE INC.,
`Petitioner,
`
`v.
`
`Zentian Limited
`Patent Owner.
`____________________
`
`Case IPR2023-00036
`Patent No. 10,839,789
`____________________
`
`
`
`DECLARATION OF DAVID ANDERSON, Ph.D. IN SUPPORT
` OF PATENT OWNER’S PRELIMINARY RESPONSE
`
`
`
`
`
`
`
`
`
`
`Case IPR2023-00036
`DECLARATION OF DAVID ANDERSON, PH.D. ISO
` PATENT OWNER’S PRELIMINARY RESPONSE
`
`
`TABLE OF CONTENTS
`
`
`Introduction
`Engagement
`Background and qualifications
`
`Relevant legal standards
`Person of ordinary skill in the art
`Burden of proof
`Claim construction
`
`1
`1
`1
`5
`5
`5
`8
`8
`8
`10
`12
`13
`13
`15
`
`I.
`A.
`B.
`C. Materials considered
`II.
`A.
`B.
`C.
`D. Obviousness
`III. Overview of the ’789 Patent
`IV. Smyth
`V. Mozer
`VI. The ’789’s requirement for “[a]n acoustic coprocessor”
`VII. The ’789 patent’s requirement that “the calculation apparatus and the
`
`acoustic model memory are fabricated on a single integrated circuit”
`
`
`
`
`
`
`
`
`
`
`
`- i -
`
`
`
`
`
` Case IPR2023-00036
`
`DECLARATION OF DAVID ANDERSON, PH.D. ISO
` PATENT OWNER’S PRELIMINARY RESPONSE
`
`
`
`EXHIBIT LIST
`
`
`Description
`
`Exhibit No.
`2001
`
`
`
`
`
`
`
`
`
`
`
`
`- ii -
`
`
`
`
`
`
`
`
`
`
`I.
`
`I, David Anderson, Ph.D, do hereby declare as follows:
`
`Introduction
`
`Engagement
`
`A.
`
`1.
`
`I have been retained by Patent Owner Zentian Limited (“Zentian” or
`
`“Patent Owner”) to provide my opinions with respect to Zentian’s Preliminary
`
`Response to the Petition in Inter Partes Review proceeding IPR2023-00036, with
`
`respect to U.S. Pat. 10,839,789. I am being compensated for my time spent on this
`
`matter. I have no interest in the outcome of this proceeding and the payment of my
`
`fees is in no way contingent on my providing any particular opinions.
`
`2.
`
`As part of this engagement, I have also been asked to provide my
`
`technical review, analysis, insights, and opinions regarding the materials cited and
`
`relied upon by the Petition, including the prior art references and the supporting
`
`Declaration of Mr. Schmandt.
`
`3.
`
`The statements made herein are based on my own knowledge and
`
`opinions.
`
`B.
`
`Background and qualifications
`
`4. My full qualifications, including my professional experience and
`
`education, can be found in my Curriculum Vitae, which includes a complete list of
`
`my publications, and is attached as Ex. A to this declaration.
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`I am a professor in the School of Electrical and Computer Engineering
`
`
`
`5.
`
`at the Georgia Institute of Technology (“Georgia Tech”) in Atlanta, Georgia. I
`
`have been a professor at Georgia Tech since 1999. In 2009 I served as a visiting
`
`professor in the Department of Computer Science at Korea University in Seoul,
`
`South Korea.
`
`6.
`
`I received my Ph.D. in Electrical and Computer Engineering from
`
`Georgia Tech in 1999. I received my B.S. and M.S. in Electrical Engineering from
`
`Brigham Young University in 1993 and 1994, respectively.
`
`7.
`
`In my employment prior to Georgia Tech as well as in my subsequent
`
`studies and research, I have worked extensively in areas related to the research,
`
`design, and implementation of speech and audio processing systems. I have also
`
`taught graduate and undergraduate level courses at Georgia Tech on the
`
`implementation of signal processing and embedded systems. For example, I have
`
`taught courses on statistical machine learning, machine learning for speech, pattern
`
`recognition, multimedia processing and systems, software design, computer
`
`architecture, real-time signal processing systems, and applications of signal
`
`processing (covering topics in audio processing and speech recognition). I have
`
`also designed and taught a course on signal processing in the context of human
`
`
`
`
`
`
`
`
`2
`
`
`
`
`
`
`
`
`
`
`
`
`perception. These courses and my research have covered many topics relevant to
`
`the subject matter of the ’789 patent and the prior art cited therein.
`
`8.
`
`I have served as principal investigator or co-principal investigator in
`
`numerous multi-disciplinary research projects including “Blind Source Separation
`
`for Audio,” “Audio Classification,” “Auditory Scene Analysis,” “Hearing Aid
`
`Audio Processing,” “Speaker Driver Sound Enhancement,” “I-Vector Based Voice
`
`Quality,” “Analysis of Voice Exercise Using Signal Processing,” and “Smart
`
`Homes for Effective and Safe Remote Work During a Pandemic and Beyond.”
`
`9.
`
`I also have extensive experience with the practical implementation of
`
`signal processing algorithms, information theory, signal detection, and related
`
`topics through my research and consulting. I have published over 200 book
`
`chapters and papers in reviewed journals and conferences. Topics include those
`
`such as “Speech recognition using filter bank features,” “Speaker adaptation using
`
`speaker similarity score on DNN features.” “Segmentation based speech
`
`enhancement using auxiliary sensors,” “A framework for estimation of clean
`
`speech by fusion of outputs from multiple speech enhancement systems,”
`
`“Distributed acquisition and processing systems for speech and audio,” “A missing
`
`data-based feature fusion strategy for noise-robust automatic speech recognition
`
`using noisy sensors,” “Learning distances to improve phoneme classification,”
`
`
`
`
`
`
`
`
`3
`
`
`
`
`
`
`
`
`
`
`
`
`“Identification of voice quality variation using i-vectors,” “Varying time-constants
`
`and gain adaptation in feature extraction for speech processing,” “Low bit-rate
`
`coding of speech in harsh conditions using non-acoustic auxiliary devices,”
`
`“Speech analysis and coding using a multi-resolution sinusoidal transform,”
`
`“Biologically inspired auditory sensing system interfaces on a chip,” “Cascade
`
`classifiers for audio classification,” and “Single acoustic channel speech
`
`enhancement based on glottal correlation using non-acoustic sensors.” I have also
`
`contributed book chapters for treatises such as “Independent Component Analysis
`
`for Audio and Biosignal Applications,” and written a book on Fixed-Point Signal
`
`Processing which is related to the practical implementation of systems for
`
`processing sound and other signals.
`
`10.
`
`I am a named inventor on eight patents, including “Speech activity
`
`detector for use in noise reduction system, and methods therefor” (U.S. Patent No.
`
`6,351,731), and “Analog audio signal enhancement system using a noise
`
`suppression algorithm” (U.S. Patent No. 7,590,250).
`
`11.
`
`I am a Senior Member of the Institute of Electrical and Electronics
`
`Engineers (“IEEE”) and have been a Member since 1991. I am also a Member of
`
`the IEEE Signal Processing Society. From 1994 to 2016, I was also a member of
`
`the Acoustical Society of America. In 2003, I served as the Co-Chair for the NSF
`
`
`
`
`
`
`
`
`4
`
`
`
`
`
`
`
`
`
`
`
`
`Symposium on Next Generation Automatic Speech Recognition. In 2004, I
`
`received the Presidential Early Career Award for Scientists and Engineers,
`
`presented by then-President George W. Bush, for my work on ultra-low-power
`
`signal processing system design.
`
`C. Materials considered
`
`12.
`
`In the course of preparing my opinions, I have reviewed and am familiar
`
`with the ’789 patent, including its written description, figures, and claims. I have
`
`also reviewed and am familiar with the Petition in this proceeding, the supporting
`
`Declaration of Mr. Schmandt, and the relied upon prior art, including Smyth and
`
`Mozer. I have also reviewed the materials cited in this declaration. My opinions are
`
`based on my review of these materials as well as my more than 30 years of
`
`experience, research, and education in the field of art.
`
`II. Relevant legal standards
`
`13.
`
`I am not an attorney. I offer no opinions on the law. But counsel has
`
`informed me of the following legal standards relevant to my analysis here. I have
`
`applied these standards in arriving at my conclusions.
`
`A.
`
`Person of ordinary skill in the art
`
`14.
`
`I understand that an analysis of the claims of a patent in view of prior
`
`art has to be provided from the perspective of a person having ordinary skill in the
`
`
`
`
`
`
`
`
`5
`
`
`
`
`
`
`
`
`
`art at the time of invention of the ’789 patent. I understand that I should consider
`
`
`
`factors such as the educational level and years of experience of those working in the
`
`pertinent art; the types of problems encountered in the art; the teachings of the prior
`
`art; patents and publications of other persons or companies; and the sophistication
`
`of the technology. I understand that the person of ordinary skill in the art is not a
`
`specific real individual, but rather a hypothetical individual having the qualities
`
`reflected by the factors discussed above.
`
`15.
`
`I understand that the Petition applies a priority date of September 14,
`
`2004, for the challenged claims, Pet. 3, and I apply the same date.
`
`16.
`
`I further understand that the Petition defines the person of ordinary skill
`
`in the art at the time of the invention as having had a master’s degree in computer
`
`engineering, computer science, electrical engineering, or a related field, with at least
`
`two years of experience in the field of speech recognition, or a bachelor’s degree in
`
`the same fields with at least four years of experience in the field of speech
`
`recognition. The Petition adds that further education or experience might substitute
`
`for the above requirements. I do not dispute the Petition’s assumptions at this time,
`
`and my opinions are rendered on the basis of the same definition of the ordinary
`
`artisan set forth in the Petition.
`
`
`
`
`
`
`
`
`6
`
`
`
`
`
`
`
`
`
`I also note, however, that an ordinarily skilled engineer at the time of
`
`
`
`17.
`
`the invention would have been trained in evaluating both the costs and benefits of a
`
`particular design choice. Engineers are trained (both in school and through general
`
`experience in the workforce) to recognize that design choices can have complex
`
`consequences that need to be evaluated before forming a motivation to pursue a
`
`particular design choice, and before forming an expectation of success as to that
`
`design choice. In my opinion, anyone who did not recognize these realities would
`
`not be a person of ordinary skill in the art. Thus, a person who would have simply
`
`formed design motivations based only on the premise that a particular combination
`
`of known elements would be possible would not be a person of ordinary skill
`
`regardless of their education, experience, or technical knowledge. Likewise, a person
`
`who would have formed design motivations as to a particular combination of known
`
`elements based only on the premise that the combination may provide some benefit,
`
`with no consideration of the relevance of the benefit in the specific context and in
`
`relation to the costs or disadvantages of that combination, would also not have be a
`
`person of ordinary skill in the art, regardless of their education, experience, or
`
`technical knowledge. In my opinion, a person of ordinary skill in the art would have
`
`been deliberative and considered, rather than impulsive.
`
`
`
`
`
`
`
`
`7
`
`
`
`
`
`
`
`
`
`18. Throughout my declaration, even if I discuss my analysis in the present
`
`
`
`tense, I am always making my determinations based on what a person of ordinary
`
`skill in the art (“POSA”) would have known at the time of the invention. Based on
`
`my background and qualifications, I have experience and knowledge exceeding the
`
`level of a POSA, and am qualified to offer the testimony set forth in this declaration.
`
`B.
`
`19.
`
`Burden of proof
`
`I understand that in an inter partes review the petitioner has the burden
`
`of proving a proposition of unpatentability by a preponderance of the evidence.
`
`C. Claim construction
`
`20.
`
`I understand that in an inter partes review, claims are interpreted based
`
`on the same standard applied by Article III courts, i.e., based on their ordinary and
`
`customary meaning as understood in view of the claim language, the patent’s
`
`description, and the prosecution history viewed from the perspective of the ordinary
`
`artisan. I further understand that where a patent defines claim language, the
`
`definition in the patent controls, regardless of whether those working in the art may
`
`have understood the claim language differently based on ordinary meaning.
`
`D. Obviousness
`
`21.
`
`I understand that a patent may not be valid even though the invention
`
`is not identically disclosed or described in the prior art if the differences between the
`
`
`
`
`
`
`
`
`8
`
`
`
`
`
`
`
`
`
`subject matter sought to be patented and the prior art are such that the subject matter
`
`
`
`as a whole would have been obvious to a person having ordinary skill in the art in
`
`the relevant subject matter at the time the invention was made.
`
`22.
`
`I understand that, to demonstrate obviousness, it is not sufficient for a
`
`petition to merely show that all of the elements of the claims at issue are found in
`
`separate prior art references or even scattered across different embodiments and
`
`teachings of a single reference. The petition must thus go further, to explain how a
`
`person of ordinary skill would combine specific prior art references or teachings,
`
`which combinations of elements in specific references would yield a predictable
`
`result, and how any specific combination would operate or read on the claims.
`
`Similarly, it is not sufficient to allege that the prior art could be combined, but rather,
`
`the petition must show why and how a person of ordinary skill would have combined
`
`them.
`
`23.
`
`I understand that where an alleged motivation to combine relies on a
`
`particular factual premise, the petitioner bears the burden of providing specific
`
`support for that premise. I understand that obviousness cannot be shown by
`
`conclusory statements, and that the petition must provide articulated reasoning with
`
`some rational underpinning to support its conclusion of obviousness. I also
`
`understand that skill in the art and “common sense” rarely operate to supply missing
`
`
`
`
`
`
`
`
`9
`
`
`
`
`
`
`
`
`
`knowledge to show obviousness, nor does skill in the art or “common sense” act as
`
`
`
`a bridge over gaps in substantive presentation of an obviousness case.
`
`III. Overview of the ’789 Patent
`
`24. U.S. Patent 10,839,789, titled “Speech recognition circuit and method,”
`
`is directed to an improved speech recognition circuit and associated methods. Ex.
`
`1001, 1:20-21. The ’789 patent teaches and claims an acoustic co-processor
`
`comprising a calculating apparatus for calculating distances between a feature vector
`
`and acoustic states of an acoustic model, which acoustic states and acoustic model
`
`are stored in and read from an acoustic model memory that is fabricated on the same
`
`integrated circuit as the calculating apparatus. Claim 1; Ex. 1001, 25:40-55, 34:56-
`
`63.
`
`25. The ’789 patent teaches that an “audio input for speech recognition”
`
`may be input to the front end in the form of digital audio or analog audio that is
`
`converted to digital audio using an analog to digital converter. Ex. 1001, 12:52-12:55
`
`“The audio input is divided into time frames, each time frame typically being on the
`
`order of 10 ms.” Ex. 1001, 12:55-57. “For each audio input time frame, the audio
`
`signal is converted into a feature vector. This may be done by splitting the audio
`
`signal into spectral components,” such as, for instance, 13 components plus their
`
`first and second derivatives, creating a total of 39 components. Ex. 1001, 12:58-60.
`
`
`
`
`
`
`
`
`10
`
`
`
`
`
`
`
`
`
`The feature vector thus “represents a point in an N-dimensional space,” where N is
`
`
`
`generally in the range of 20 to 39. Ex. 1001, 13:22-23.
`
`26. Each feature vector is then passed to the calculating circuit, or distance
`
`calculation engine, which calculates a distance indicating the similarity between a
`
`feature vector and one or more predetermined acoustic states of an acoustic model.
`
`Ex. 1001, 6:63-65, 25:42-44 (“Each feature vector is transferred to a distance
`
`calculation engine circuit 204, to obtain distances for each state of the acoustic
`
`model.”). “The distance calculator stage of the recognition process computes a
`
`probability or likelihood that a feature vector corresponds to a particular state.” Ex.
`
`1001, 13:26-29. “The likelihood of each state is determined by the distance between
`
`the feature vector and each state.” Ex. 1001, 13:3-4. The distance calculation may
`
`be a Mahalanobis distance using Gaussian distributions. Ex. 1001, 4:23-24. “The
`
`MHD (Mahalanobis Distance) is a distance between two N-dimensional points,
`
`scaled by the statistical variation in each component.” Ex. 1001, 13:15-18. The ’789
`
`patent teaches calculating the distance between a feature vector and 8,000 states,
`
`“i.e. one distance for each of the 8,000 states,” Ex. 1001, 13:61-63, which it teaches
`
`“gives the best recognition results when used with a language model.” Id. at 13:10-
`
`15, 14:5-6 (“Each state is also a 39 dimensional vector, having the same spectral
`
`components as the feature vector.”). “Due to the 10 ms frame length, a feature vector
`
`
`
`
`
`
`
`
`11
`
`
`
`
`
`
`
`
`
`arrives at the MHD engine,” i.e., the distance calculation engine or calculating
`
`
`
`circuit, “every 10 ms.” Ex. 1001, 14:1-2.
`
`27. The distance calculation engine or calculating circuit “may be included
`
`within an accelerator,” Ex. 1001, 3:62-64, which may be a “loosely bound co-
`
`processor for a CPU running speech recognition software,” and which “has the
`
`advantage of reducing computational load on the CPU, and reducing memory
`
`bandwidth load for the CPU.” Id. at 24:26-31; see Figs. 17-23. “Each time a feature
`
`vector is loaded into the accelerator, the accelerator computes the distances for all
`
`states for that feature vector[.]” Ex. 1001, 26:15-18.
`
`28.
`
`“The distances calculated by the distance calculation engine are then
`
`transferred to the search stage 106 of the speech recognition circuit, which uses
`
`models such as one or more word models and/or language models to generate and
`
`output recognised text.” Ex. 1001, 23:14-19.
`
`IV. Smyth
`
`29. U.S. Pat. 5,819,222, titled “Task-constrained connected speech
`
`recognition of propagation of tokens only if valid propagation path is present,”
`
`(“Smyth”), is directed to “task-constrained connected word recognition where the
`
`task, for example, might be to recognise one of a set of account numbers or product
`
`codes.” Ex. 1005, 1:15-21.
`
`
`
`
`
`
`
`
`12
`
`
`
`
`
`
`
`
`
`
`
`
`V. Mozer
`
`30. U.S. Pat. 6,832,194, titled “Audio recognition peripheral system,”
`
`(“Mozer”), is directed to “integrated circuits for implementing audio recognition,”
`
`offering a “low cost audio recognition peripheral to operate in conjunction with a
`
`processor.” Ex. 1046, 1:5-10, 2:4-7.
`
`VI. The ’789’s requirement for “[a]n acoustic coprocessor”
`
`
`31. Every challenged claim of the ’789 patent recites “[a]n acoustic
`
`coprocessor.”
`
`32. The ’789 patent teaches that in the relevant embodiment, “the distance
`
`calculation engine is designed as a speech accelerator, to operate as a loosely
`
`bound co-processor for a CPU running speech recognition software. This has the
`
`advantage of reducing the computational load on the CPU, and reducing memory
`
`bandwidth load for the CPU.” Ex. 1001, 24:26-31 (emphasis added).
`
`33. The ’789 patent explains that the claimed “coprocessor” (1) performs
`
`the function of the “accelerator” taught throughout the specification; (2) is not the
`
`main CPU; and (3) must work in conjunction with a main CPU. The ’789 patent’s
`
`teachings as to the claimed “coprocessor” in those respects are consistent with
`
`contemporaneous dictionary definitions, which define “coprocessor” as “a
`
`processor that is connected to a main processor and operates concurrently with
`
`
`
`
`
`
`
`
`13
`
`
`
`
`
`
`
`
`
`the main processor,” Ex. 2003 at 3 (emphasis added), and as “[a] processor used in
`
`
`
`conjunction with a central processing unit. . . .” Ex. 2009 at 3.
`
`34. The ’789 patent’s teachings as to the “accelerator” are likewise
`
`consistent with the above definitions and teachings, consistently disclosing that the
`
`“accelerator” must work in conjunction with another processor that is the main
`
`CPU. See, e.g., Ex. 1001, Figs. 18-22, 26:44-46 (“The Acoustic Model may be
`
`loaded into the Acoustic Model Memory by software running on the CPU prior to
`
`the first use of the Accelerator.”); 32:11-16 (“the operation of the CPU and
`
`accelerator are tightly coupled”). There are no embodiments to the contrary.
`
`35. The Petition identifies Smyth’s classifier 34 and at least one alleged
`
`signal line as the claimed “coprocessor.” Pet. 11, 16, 44. The Petition, citing to Mr.
`
`Schmandt’s declaration at paragraph 129, alleges that Smyth’s classifier and the
`
`alleged associated signal lines “are collectively a coprocessor because the
`
`classifier is designed to operate with the feature extractor and the sequencer (and
`
`other components of speech recognizer 3) to calculate distances and performed
`
`[sic] other claimed functionality.” Pet. 16.
`
`36. The Petition’s showing falls short of meeting the basic and well-
`
`known meaning of “coprocessor,” as taught in the ’789 patent and
`
`contemporaneous dictionaries. In particular, the Petition’s theory as to
`
`
`
`
`
`
`
`
`14
`
`
`
`
`
`
`
`
`
`
`
`
`“coprocessor” as set forth with respect to independent claims 1 and 10 never
`
`demonstrates (or even alleges) that (1) Smyth’s classifier 34 is not the main CPU;
`
`and (2) that Smyth’s classifier 34 works in conjunction with another processor that
`
`is the main CPU.
`
`37.
`
` The Petition’s allegation that the classifier 34 works with “the feature
`
`extractor and the sequencer (and other components of speech recognizer 3)” is not
`
`evidence that Smyth’s classifier 34 is not the main CPU or that it works in
`
`conjunction with a main CPU, as the definition of “coprocessor” requires. A
`
`processor that simply works with other processors and other components of a
`
`speech recognition system is not necessarily a coprocessor, because such a
`
`processor may in fact be the main CPU, or else may not work in conjunction with a
`
`main CPU at all.
`
`VII. The ’789 patent’s requirement that “the calculation apparatus and the
`acoustic model memory are fabricated on a single integrated circuit”
`
`
`
`38.
`
`I understand independent claim 1 and its dependents, as well as claims
`
`11 and 24 require that the claimed “calculating apparatus” and the claimed
`
`“acoustic model memory” are “fabricated on a single integrated circuit.”
`
`39. The Petition initially relies on Smyth alone to meet those
`
`requirements. Pet. 38-40. According to the Petition, Smyth’s “calculating
`
`
`
`
`
`
`
`
`15
`
`
`
`
`
`
`
`
`
`
`
`
`apparatus” is the classifier processor 341, and Smyth’s “acoustic model memory”
`
`is the state memory 342. Pet. 38.
`
`40. Smyth teaches that the classifier processor 341 and the state memory
`
`342 are distinct structures, and does not depict or otherwise describe them as
`
`being “fabricated on a single integrated circuit.” Ex. 1005, Fig. 3, 5:54-6:7.
`
`
`
`41. The Petition theorizes that because the processor 341 could be a
`
`Motorola DSP56000, and because the Motorola DSP56000 had on-board memory,
`
`“Smyth teaches or renders obvious a single integrated circuit (IC) on which is
`
`fabricated both the classifier processor 341 and state memory 342.” Pet. 38-40
`
`(emphasis added). The Petition provides and relies on Kloker, Ex. 1009, which
`
`describes the design details of the Motorola DSP56000.
`
`16
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`42. The combined teachings of Kloker and Smyth, however, definitively
`
`prove that the onboard memory of the Motorola DSP56000 could not be the state
`
`memory 342 of Smyth, and could not serve as the recited “acoustic model
`
`memory” of the challenged claims.
`
`43. Kloker teaches that the Motorola DSP56000 had a total memory size
`
`of only 512 words (24-bit) of RAM, as shown in Kloker’s Fig. 2 under the labels
`
`“x memory” and “y memory” (256 words of RAM in each memory). Ex. 1009 at 4
`
`(Fig. 2) (annotated).
`
`
`
`
`
`
`
`
`17
`
`
`
`
`
`
`
`
`
`
`
`Since bytes are 8-bits, each of Kloker’s 24-bit words is 3 bytes. Thus, Kloker’s 512
`
`
`
`words of memory only amounts to approximately 1.5 kilobytes that could even
`
`theoretically be used for storing data structures.
`
`44.
`
` While the Motorola DSP56000 also has 2048 words (24-bit) of
`
`program ROM, that ROM memory is read-only memory used for storing
`
`executable code and defining the processor operation. The ROM blocks in the “x
`
`memory” and “y memory” are likewise programmed with data used for Fourier
`
`transforms. These are not modifiable by a programmer, developer, or other user of
`
`the DSP processor. Accordingly, the ROM memory could not be used to store the
`
`recited “acoustic model data.” Nonetheless, even the entire ROM memory is only
`
`approximately 7.5 kilobytes in size, and the entirety of the Motorola DSP56000’s
`
`memory is barely 9 kilobytes.
`
`45. By contrast, Smyth teaches that the state memory 342 holds multiple
`
`“state field[s]” for each of a “plurality of speech states,” e.g., three states per
`
`allophone. Ex. 1005, 5:58-62. As explained below, an ordinary artisan would have
`
`understood from that teaching that Smyth’s state memory 342 must have at least
`
`750 kilobytes of data storage capacity—480x more than the usable RAM in the
`
`Motorola DSP56000, and 80x times more than the combined ROM and RAM in
`
`that device.
`
`
`
`
`
`
`
`
`18
`
`
`
`
`
`
`
`
`
`46. An ordinary artisan would have been able to calculate the approximate
`
`
`
`storage requirement of Smyth’s state memory 342 because Smyth discloses using
`
`what a POSITA would recognize as triphones having three states per allophone,
`
`Ex. 1005, 5:58-62, of which there are tens of thousands. The ordinary artisan
`
`would have known that even in a resource-constrained system, realistically at least
`
`2000 allophones (triphones) of the more common allophones are used. Each
`
`acoustic state requires storage of the mean and variance for each Gaussian for each
`
`feature in the vector which may be of dimension 16-40 (Smyth references a paper
`
`that uses 16 features per vector but more features is more common). Ex. 1005,
`
`5:31-45 (citing “On the Evaluation of Speech Recognisers and Databases using a
`
`Reference System”, Chollet & Gagnoulet, 1982 proc. IEEE p2026 and “On the use
`
`of Instantaneous and Transitional Spectral Information in Speaker Recognition”,
`
`Soong & Rosenberg, 1988 IEEE Trans. On Acoustics, Speech and Signal
`
`Processing Vol. 36 No. 6 p871). Using 16 to 40 features per vector would
`
`require 128-320 bytes of memory. Thus, if 2000 allophones are used, with three
`
`states each (as disclosed by Smyth and commonly used in ASR), the acoustic
`
`model would require more than 750 kBytes of memory. This estimate is on the low
`
`end. At the time of the patents in question, typical acoustic models had states
`
`represented by multiple Gaussians (Gaussian mixtures) and higher-dimensional
`
`
`
`
`
`
`
`
`19
`
`
`
`
`
`
`
`feature vectors to have acceptable performance. As the’789 patent itself teaches,
`
`such models may occupy many megabytes of storage space. (See Ex. 1001, 34:52-
`
`
`
`
`
`
`66).
`
`47. Accordingly, the 1.5kB RAM onboard memory of the Motorola
`
`DSP56000 (or even the entire 9-kilobyte RAM and ROM memory) could not be
`
`the state memory 342 taught in Smyth, and also could not be an “acoustic model
`
`memory” as recited in the claims.
`
`48.
`
`It necessarily follows that Smyth’s state memory 342 was not
`
`fabricated on the same integrated circuit as the Motorola DSP56000, since the
`
`design of the integrated circuit of the Motorola DSP56000 is shown in Kloker, and
`
`that design does not include any memory that meets the requirements of the state
`
`memory 342. Smyth therefore did not teach or suggest that its state memory 342
`
`and processor 341 were fabricated on a single integrated circuit, as claims 1-9 and
`
`11 and 24 require.
`
`49. The Petition alternatively proposes a combination of Smyth and
`
`Mozer. Pet. 41-44.
`
`50. Mozer’s relevant embodiment teaches a memory 460 consisting of a
`
`4K SRAM. Ex. 1046, 10:20-21. Mozer does not specific whether its 4K SRAM has
`
`4K bits (which would be roughly 512 bytes or .5 kB) or 4K bytes (which would be
`
`
`
`
`
`
`
`
`20
`
`
`
`
`
`
`
`
`
`4kB). But even assuming generously that Mozer’s RAM had 4 kB of storage space,
`
`
`
`that memory size would have been far too small to serve as Smyth’s state memory
`
`342, which would have required more than 750 kB, as explained above. Indeed, in
`
`the sections of Mozer cited by the Petition, Mozer only teaches that the memory
`
`460 holds “a first vector representation of the audio signal 425 and a second vector
`
`representing a template.” Ex. 1046, 10:17-20. That is a tiny fraction of the
`
`information that Smyth teaches as being stored in its state memory 342. Thus,
`
`Mozer could not have taught or suggested to an ordinary artisan the idea of
`
`fabricating Smyth’s state memory 342 and its classifier processor 341 on a single
`
`integrated circuit.
`
`51. The Petition’s stated motivation for modifying Smyth “to form the
`
`classifier processor and associated memory on a single IC” could not possibly have
`
`motivated an ordinary artisan to undertake such a modification. The Petition’s
`
`articulated motivation is that modifying Smyth in that manner would have allowed
`
`Smyth’s recognition system to “offload” vector processing operations associated
`
`with audio recognition to Smyth’s audio recognition peripheral, i.e., Smyth’s
`
`classifier 34. Pet. 42. Smyth already offloads vector processing operations
`
`associated with audio recognition to its classifier 34 without the Petition’s
`
`proposed modification. Pet. 42 (“Smyth already teaches the classifier performing
`
`
`
`
`
`
`
`
`21
`
`
`
`
`
`
`
`
`
`
`
`
`the distance calculations is performed by a processor distinct from the processor
`
`for performing feature vector calculation and the processor for performing word
`
`identification.”).
`
`52. An ordinary artisan likewise would have known that Smyth, without
`
`modification, already achieves the exact benefit the Petition contends the ordinary
`
`artisan would have sought to achieve by modifying Smyth. Accordingly, the
`
`Petition’s articulated motivation could not have possibly motivated the ordinary
`
`artisan to “fabricate classifier processor 341 . . . and state memory 342 . . . on a
`
`single integrated circuit.” Pet. 44.
`
`53. The Petition’s proposed modification requires fabricating an
`
`integrated circuit. Pet. 42. Fabricating a circuit would have been well beyond the
`
`skill level of the Petition’s stated person of ordinary skill. According to the
`
`Petition, the person of ordinary skill in this context would have had a master’s
`
`degree in computer engineering, computer science, electrical engineering, or a
`
`related field, along with two to four years of work experience in “speech
`
`recognition.” Pet. 3. It is my opinion that a person with those qualifications would
`
`not have been able to undertake the highly complex task of fabricating a new
`
`integrated circuit.
`
`
`
`
`
`
`
`
`22
`
`
`
`
`
`
`
`
`
`
`
`
`54. The Petition’s assertion that others in the past had succeeded in
`
`fabricating integrated circuits, even those allegedly with a calculating apparatus
`
`and an acoustic model memory, Pet. 43-44 (citing Toyoda), would have been
`
`irrelevant to the ordinary artisan’s expectation of success because the ordinary
`
`artisan as defined by the Petition did not have the capability to achieve that feat.
`
`Accordingly, the ordinary artisan would not have had a reasonable expectation of
`
`success with respect to the Petition’s proposed modification.
`
`55.
`
`I hereby declare that all statements made herein of my own
`
`knowledge are true and that all opinions expressed herein are my own; and further
`
`that these statements were made with the knowledge that willful false statements
`
`and the like are punishable by fine or imprisonment, or both, under Section 1001 of
`
`Title 18 of the United States Code.
`
`
`
` ______________________________
`Executed on March 15, 2023
`
`David Anderson, Ph.D
`
`
`
`
`
`
`
`
`
`23
`
`
`
`
`
`
`
`
`
`
`
`
`
`
`School of Electrical and Computer Engineering
`Georgia Institute of Technology
`(cid:72) 770-883-0708
`(cid:66) anderson@gatech.edu
`
`David V. Anderson
`March 1, 2023
`Earned Degrees
`1999 Ph.D. in Electrical and Co
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
