`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`
`________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`
`________________
`
`ULTRATEC, INC.
`Petitioner
`
`v.
`
`SORENSON IP HOLDINGS, LLC
`Patent Owner
`________________
`
`Case IPR2017-01394
`________________
`
`Patent No. 9,336,689
`
`
`
`DECLARATION OF BENEDICT J. OCCHIOGROSSO IN SUPPORT OF
`PATENT OWNER’S PRELIMINARY RESPONSE
`
`
`
`
`
`
`
`
`
`
`
`
`
`1
`
`
`
`
`
`
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 1 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`
`Declaration of Benedict J. Occhiogrosso
`
`I, Benedict J. Occhiogrosso, declare as follows:
`
`1.
`
`I make this declaration based upon my own personal knowledge and,
`
`if called upon to testify, would testify competently to the matters contained herein.
`
`2.
`
`I have been asked to provide technical assistance in the above-
`
`captioned inter partes review (“IPR”) proceeding (Case IPR2017-01394).
`
`3.
`
`This declaration is a statement of my opinions on issues related to the
`
`patentability of claims 1-12 and 19 of U.S. Patent No. 9,336,689 (the “’689
`
`Patent”) (Ex. 1003).
`
`I.
`
`Background and Qualifications
`
`4. My qualifications are stated more fully in my curriculum vitae. Ex.
`
`2002. Here, I provide a brief summary of my qualifications:
`
`5.
`
`I hold a Bachelor of Science Degree in Electrical Engineering as well
`
`as a Master of Science Degree in Electrical Engineering, both from the Polytechnic
`
`Institute of Brooklyn.
`
`6.
`
`I have authored or co-authored nearly three dozen articles in peer-
`
`reviewed journals, conference proceedings, texts, industry trade publications, and
`
`monographs. These publications span a range of topics including: Integrated
`
`
`
`-2-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 2 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`Voice–Data Communications/Switching, Integrated Packet-Circuit Switching,
`
`Voice Digitization, Packet Voice, Indoor Wireless distribution, Disaster Recovery
`
`and Business Continuity, Data Center Engineering, Switching Processor
`
`Architecture, Telephone and Voice Mail Systems, PBX & LAN switching
`
`premises-based systems and related technologies and Internet of Things (IoT).
`
`7.
`
`I have more than 40 years of telecommunications and information
`
`technology experience.
`
`
`
`I am
`
`the co-founder and President of DVI
`
`Communications Inc., a telecommunications and information technology and
`
`business consulting firm. Since the establishment of DVI in 1979, I have planned,
`
`designed, implemented, and managed large scale projects involving wired and
`
`wireless communications systems, which included transmission of voice and data.
`
`Prior to founding DVI and for several years thereafter, I held a Department of
`
`Defense security clearance and worked on several classified programs within the
`
`defense industry, where I supported the development of several pioneering
`
`technologies that have served as the prototypes for many telecommunications and
`
`IT systems later utilized in commercial practice.
`
`8.
`
`I have served in numerous leadership positions including Project and
`
`Program Manager for both research & development and consulting projects for
`
`various organizations including: U.S. Department of Defense’s Defense Advanced
`
`Research Projects Agency (DARPA); the Defense Communications Agency;
`
`
`
`-3-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 3 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`Xerox; United Technologies - Norden Systems; AT&T; Citicorp; Con-Edison and
`
`various financial services, healthcare, utility and educational institutions. I also
`
`headed an Engineering team to develop a voice mail product, which was
`
`successfully integrated and commercially produced with an off-the-shelf PBX
`
`manufacturer’s product offerings.
`
`9.
`
`I have extensive expertise in voice-data video switching, transmission
`
`systems deployed in networks, including both circuit switching and packet
`
`switching using wireline and wireless distribution methods (including Land Mobile
`
`radio, Satellite, microwave, cellular and Wi-Fi). In addition, I have developed
`
`various applications systems including voicemail, e-mail, unified messaging, and
`
`audio/video recording for a variety of facility types including call-contact centers,
`
`data centers, trading floors, and mission-critical communications centers. I have
`
`detailed knowledge of Internet Protocol (IP) technology in general and also Voice
`
`over Internet Protocol (VoIP). I have extensive expertise in voice digitization,
`
`buffering, relay and delay techniques, as well as digital signal processing of voice
`
`signals. At present, my primary responsibilities encompass strategic planning and
`
`systems design of client IT Infrastructures and program management for major
`
`projects undertaken by DVI.
`
`10.
`
`I have also developed
`
`technical expertise with
`
`regard
`
`to
`
`telecommunication applications utilizing speech recognition, including call center
`
`
`
`-4-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 4 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`and interactive voice response applications. I have supported numerous enterprise
`
`and institutional clients in the design of their call centers. My experience in this
`
`field includes the use of both speaker-independent and speaker-dependent speech
`
`recognition algorithms
`
`in different applications utilizing
`
`isolated speech
`
`recognition and continuous speech recognition.
`
`11.
`
`I have also worked on behalf of the Metropolitan Transportation
`
`Authority (MTA) on relay services and assistive technologies. For example, I
`
`worked on voice processing for captioning of live meeting proceedings, which are
`
`then made available for the deaf and hard of hearing. I have also worked on
`
`captioning or transcription of previously recorded information for compliance and
`
`record-keeping purposes. Additionally, I have designed a telecommunication relay
`
`service for multilingual translation for the United Nations, and have also deployed
`
`audio conferencing systems with transcription services for various member firms
`
`in the financial services community supporting multiple applications. (e.g.,
`
`investor conference, morning call, and alert purposes).
`
`12. As mentioned above, on behalf of the United Nations, I designed a
`
`telecommunications relay service for multilingual translation. The project
`
`included design of a system that allowed the audio from a representative to be
`
`routed to multiple translators who could repeat what was being said in other
`
`languages. The translated speech channels could then be transmitted to and
`
`
`
`-5-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 5 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`accessed by individuals in real-time who needed language assistance to understand
`
`the speaker. Non–real time transcriptions of the audio were subsequently
`
`generated into proceedings in the UN’s official languages.
`
`13. Also, as mentioned above, on behalf of the MTA, I implemented a
`
`system for captioning live meetings’ proceedings for the deaf and hard of hearing
`
`in compliance with the Americans with Disabilities Act (ADA). The project
`
`included logistics, audio routing, voice processing, video display and captioning to
`
`generate a transcription of a speaker’s voice at a public meeting, and display of the
`
`resulting transcription on television monitors in the boardroom as well as over
`
`institutional broadcast channels and Internet feeds. In–line corrections would be
`
`generated and displayed by the transcriptionist during the proceedings.
`
`14. Furthermore, I have over 40 years of experience working with and
`
`developing
`
`the constituent
`
`technologies
`
`to
`
`the ’689 Patent’s
`
`invention:
`
`telecommunication services and associated devices, speech recognition technology,
`
`telephone call routing and switching (both plain old telephone system (“POTS”)
`
`and IP-based telephony), the transmission of voice and data over POTS lines,
`
`digital transmission facilities (wired and wireless) or IP-based communications,
`
`digitization of voice, relay systems, assistive technologies for the deaf or hard of
`
`hearing, and error detection and error correction technologies (from transmission,
`
`format and up to content levels) for both real time and non-real time applications.
`
`
`
`-6-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 6 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`15. For example, with telecommunication services and associated devices
`
`for the switching and transmission of voice and data over POTS lines, digital
`
`transmission facilities (such as T1, E1 and ISDN PRI and BRI), cellular voice and
`
`data and other wireless–based technologies (Line of Sight Microwave, Land
`
`Mobile Radio, Satellite) and IP- and VoIP-based communications, I have deployed
`
`various types of switching and routing systems (supporting various voice and data
`
`communications networks) at major institutional clients such as Citicorp NY, City
`
`University of New York, Bear Stearns, Societe Generale, RPCI and Bertelsmann.
`
`See Ex. 2002 at 2-5. For these projects, among other things, I had to understand
`
`the technical implications, shortcomings, and advantages of deploying different
`
`types of switching systems interconnecting voice and data channels, and the
`
`manner in which various communication protocols transmit voice and data and
`
`would use available bandwidth on different transmission facilities.
`
` The
`
`transmission facilities included both terrestrial (wired and wireless) as well as
`
`satellite-based communications. I also had to understand the compatibility issues
`
`with various end-user devices and central network components, both at my client’s
`
`locale as well as with the telephone provider/network provider’s switching
`
`systems. This encompassed both signaling and switching protocols and interfaces
`
`to support compatibility and interoperability.
`
`
`
`-7-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 7 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`16. With respect to speech recognition technology, I have worked on
`
`numerous projects that incorporated speech recognition technology in interactive
`
`voice response, dictation, and messaging systems. For example, some of my
`
`clients used speech recognition software to convert voice messages into text that
`
`can be sent to the recipient via email or a text message (so-called unified
`
`messaging systems). For numerous clients, I deployed speech recognition
`
`technology into Interactive Voice Response (IVR) systems, improving the user
`
`experience as well as (in many instances) expediting the process flow for various
`
`real-time inquiry/response and transaction processing applications. For another of
`
`my clients, I worked on refining speech recognition protocols to be used in their
`
`telecommunications system for directory assistance. Other applications included
`
`use of speech recognition for medical transcription as well as record–keeping and
`
`compliance purposes.
`
` I applied multiple speaker-dependent and speaker-
`
`independent products.
`
`17. As an example of telephone call routing and switching, I have worked
`
`on countless PBX systems (TDM and VoIP) and other business-wide voice and
`
`data communications (transmission and switching) technologies. These projects
`
`have often included thousands of extensions deployed across multiple physical
`
`locations. For these projects, I employed detailed knowledge of the switching and
`
`signaling protocols and products available for routing POTS and VoIP calls among
`
`
`
`-8-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 8 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`the various end-user locations. Additionally, as described above, the implications
`
`and interaction of voice and data on bandwidth and call-routing across these
`
`multiple locations and numerous extensions was an important consideration.
`
`18.
`
`I also have extensive experience related to the digitization of voice.
`
`For many of my projects related to call routing and switching, calls can come into
`
`the systems as digital communications or as analog POTS calls. I have frequently
`
`designed and implemented systems that can detect the digital or analog nature of
`
`the incoming voice and when analog, utilize an appropriate sample rate and codec
`
`to convert the analog voice into digital transmission. Under the same umbrella, I
`
`have designed and implemented voice messaging systems that digitally record an
`
`analog voice message in order to record or perform other digital processing on the
`
`voice signal as described above. For example, for DARPA (Defense Advanced
`
`Research Projects Agency), I performed comparisons among various voice
`
`digitization algorithms’ performance in the presence of noise and jamming. As
`
`another example, for both DARPA and Defense Communications Agency (DCA),
`
`I conducted studies comparing circuit, packet and hybrid or
`
`integrated
`
`(circuit/packet) switching systems of voice and data traffic comparing them
`
`relative
`
`to
`
`cost,
`
`expandability,
`
`complexity,
`
`performance
`
`and
`
`reliability/survivability. As part of typical reliability considerations, I would
`
`deploy manual and automatic recovery processes which provided fault – tolerant
`
`
`
`-9-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 9 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`and resilient voice and data networking offering graceful degradation in the
`
`presence of errors or failures.
`
` Techniques employed encompassed error
`
`detection/retransmission, vigilant monitoring of correct operation coupled with
`
`switchover to backup, bypass and/or rerouting around failed switching and
`
`transmission facilities and reconnection.
`
`19. As described above, I have experience developing and implementing
`
`captioning communication systems and methods and have worked to develop
`
`captioning systems, methods, and devices. In addition, I have served as an expert
`
`in many legal and administrative proceedings involving text captioning systems
`
`involving both inline and traditional captioning.
`
`20. As previously outlined, I have supported the deployment of several
`
`systems for generating transcriptions of previously recorded and real–time audio.
`
`For example, to support a medical dictation application, I deployed a system that
`
`used speech recognition to generate transcripts for medical dictation, both manual
`
`correction and auto text correction, of medical terms that were used, to correct the
`
`medical practitioner’s dictation. For ADA-compliant broadcast of textual
`
`messages for the deaf and hard of hearing public at MTA meetings, I designed,
`
`selected, and deployed remote text captioning of meetings with remote in–line
`
`manual correction of errors in real-time. I also assessed captioning service
`
`accuracy and latency with the goal of improving service levels.
`
`
`
`-10-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 10 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`21. As described above, I also have extensive experience developing and
`
`implementing error correction systems and methods and have worked to develop
`
`systems and methods for error correction throughout my career. In addition, I have
`
`served as an expert in several legal and administrative proceedings involving error
`
`correction systems and methods. As described above, for the MTA, I worked on
`
`voice processing for captioning of live meeting proceedings, which are then made
`
`available for the deaf and hard of hearing. I have also worked on captioning or
`
`transcription of previously recorded information for compliance and record-
`
`keeping purposes. Additionally, I have designed a telecommunication relay
`
`service for multilingual translation for the United Nations, and have also deployed
`
`audio conferencing systems with transcription services for various member firms
`
`in the financial services community supporting multiple applications. (e.g.,
`
`investor conference, morning call, and alert purposes). I have also used speech
`
`recognition for medical transcription as well as record–keeping and compliance
`
`purposes. Finally, I have served as an expert for CaptionCall in validity and
`
`infringement proceedings
`
`involving U.S. Patent No. 8,379,801,
`
`including
`
`IPR2013-00288 in which the Board found that I was “qualified to testify as to
`
`matters concerning text caption communications systems, including ‘the electronic
`
`generation, correction, and display of transcribed or captioned text that is
`
`transmitted to and displayed on an electronic device.’” See Ex. 1011 at 27.
`
`
`
`-11-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 11 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`
`II. Legal Understanding
`
`
`22. My opinions are also informed by my understanding of the relevant
`
`law. I understand that patentability analysis is conducted on a claim-by-claim
`
`basis and that there are several possible reasons that a patent claim may be found
`
`unpatentable.
`
`A. Anticipation
`
`21.
`
`I understand that for a patent claim to be anticipated, each and every
`
`element of the claim must be disclosed, either expressly or inherently, in a single
`
`prior art reference, and that the prior art reference must disclose all of the elements
`
`of the claim arranged in the same way as the claim.
`
`22.
`
`I understand that for the prior art to anticipate a claim when a claim
`
`element is missing from the prior art, the prior art reference must make clear that
`
`the missing claim element is necessarily present in the prior art reference, and that
`
`it would be so recognized by persons of ordinary skill.
`
`23.
`
`I understand that to anticipate a claim, the prior art reference must
`
`enable the person of ordinary skill in the art to make and use the invention as
`
`claimed. The reference must put those skilled in the art in possession of the
`
`invention as claimed.
`
`B. Non-obviousness
`
`
`
`-12-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 12 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`24.
`
`I understand that in order to invalidate a patent claim as obvious, there
`
`must be evidence that the claimed invention as a whole would have been obvious
`
`to a person of ordinary skill in the relevant art at the time the invention was made.
`
`25.
`
`I understand that the non-obviousness of a claim over the prior art is a
`
`question of law based on the following underlying factual inquiries: (a) the scope
`
`and content of the prior art, (b) ascertaining the differences between the claimed
`
`invention and the prior art, (c) the level of ordinary skill in the art, and (d)
`
`secondary considerations (also called objective indicia) of non-obviousness.
`
`26.
`
`I understand that an invention may not be obvious when there is a
`
`suggestion to explore a new technology or general approach that seemed to be a
`
`promising field of experimentation, but where the prior art references gave only
`
`general guidance as to the particular form of the claimed invention or how to
`
`achieve it.
`
`27.
`
`I understand that a petitioner must specify where each element of the
`
`patent claim is found in the prior art. It is the petitioner’s burden to explain the
`
`prior art evidence and not the Board’s responsibility to search through the prior art
`
`references to support allegations of anticipation or obviousness. I also understand
`
`that the Board resolves vagueness and ambiguity in the petition against the
`
`petitioner, and that petitions that fail to specify where each element is found in the
`
`prior art or fail to explain the cited evidence may be denied by the Board.
`
`
`
`-13-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 13 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`28.
`
`I understand that it is wrong to use hindsight in evaluating
`
`obviousness. For example, it is wrong to use the ’689 Patent as a guide to combine
`
`the disclosures in the prior art in the right way so as to achieve the result of the
`
`’689 Patent claims. Instead, I understand that there must be a motivation as to why
`
`the person of ordinary skill in the art would combine the prior art references to
`
`arrive at the claimed invention. It is important to identify why the person of
`
`ordinary skill in the art would combine the prior art references, not just that they
`
`could be combined. I also understand that the person of ordinary skill in the art
`
`would need to have a reasonable expectation of success when combining such
`
`references. Specifically, the requirement that a person of ordinary skill in the art
`
`have a reasonable expectation of success refers to the likelihood of success when
`
`combining references to meet the limitations of the claimed invention. The inquiry
`
`is not whether one would expect the prior art references to operate as intended
`
`once combined. Instead, one must have a motivation to combine the references
`
`and an expectation of achieving what is claimed in the patent-in-suit.
`
`C.
`
`Person Of Ordinary Skill In The Art
`
`29.
`
`I understand that the issues related to whether the claims of the ’689
`
`Patent are valid in light of the prior art are considered from the perspective of what
`
`would have been known or understood by someone of ordinary skill in the art at
`
`the time of the invention.
`
`
`
`-14-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 14 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`30.
`
`I understand that the priority date of the ’689 Patent is November 24,
`
`2009. Based upon my education and experience, I am very familiar with the level
`
`of relevant knowledge about the technology at issue that one of ordinary skill in
`
`the art would have possessed as of November 24, 2009.
`
`31. As an engineer and consultant who has been heavily involved with the
`
`telecommunication industry, I was involved with the design and implementation of
`
`various telecommunication systems, speech recognition technologies, and error
`
`correction methods used in 2009, and I am personally familiar with the state of the
`
`art in the technological field of the ’689 Patent around November 24, 2009.
`
`32. The specification of
`
`the ’689 Patent describes
`
`the relevant
`
`technological field as follows: “Embodiments of the disclosure relate generally to
`
`text captioning and more specifically to correction of errors within a text caption.”
`
`Ex. 1003 at 1:16-18. In IPR2013-00288, the Board considered U.S. Patent No.
`
`8,379,801 (the “’801 Patent”) and determined that “the problem addressed by the
`
`’801 patent is correcting errors in a transcript displayed in a text caption
`
`communication system, regardless of whether the prospective recipient of the
`
`corrected transcript is deaf or hearing-impaired.” Ex. 1011 at 9. The ’801 Patent
`
`shares a specification with the ’689 Patent (Ex. 1003 at (63)) and addresses the
`
`same general concept. For purposes of this proceeding, I have adopted the Board’s
`
`definition from IPR2013-00288, with the additional consideration that someone
`
`
`
`-15-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 15 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`with less technical education but more experience or more technical education but
`
`less experience would also have been a person of ordinary skill in the art.
`
`III. The ’689 Patent
`
`33. The “’689 Patent,” entitled “Methods and Apparatuses Related to Text
`
`Caption Error Correction,” was filed on October 31, 2014 and issued on May 10,
`
`2016.
`
`34. The technology of the ’689 Patent relates “generally to text
`
`captioning.” Ex. 1003 at 1:16-17. It includes 20 claims (2 independent claims and
`
`18 dependent claims) addressed to systems and methods designed for “correction
`
`of errors within a text caption.” Ex. 1003 at 1:17-18.
`
`35. The specification of the ’689 Patent describes a communication
`
`system “configured to facilitate a communication session between a hearing-
`
`impaired user and a hearing-capable user” (Ex. 1003 at 3:40-42) as depicted in
`
`Figure 1.
`
`
`
`-16-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 16 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`
`
`Ex. 1003 at Figure 1. Specifically, the system provides for communication
`
`“between a hearing-impaired user 140 and a hearing-capable user 160.” Ex. 1003
`
`at 3:66-67. When a hearing-impaired user utilizes this system, “hearing-capable
`
`user 160 may interact in a conventional manner with communication device 120,”
`
`(Ex. 1003 at 4:2-3) while “[t]he voice of hearing-capable user 160 may be . . .
`
`transmitted over network 180 to communication device 120.” Ex. 1003 at 4:5-8.
`
`The hearing-capable user’s voice may then “be transmitted through communication
`
`device 120, over network 170, and to relay service 110.” Ex. 1003 at 4:9-10. The
`
`“[h]earing-impaired user 140 may interact in a conventional manner with
`
`communication device 190 through the use of a voice-based dialogue conveyed
`
`
`
`-17-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 17 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`over communication device 120.”
`
` Ex. 1003 at 4:14-17.
`
` Additionally,
`
`“communication device 120 may be configured to receive and display a text
`
`transcription of a voice signal sent from relay service 110 via network 170.” Ex.
`
`1003 at 4:18-21.
`
`36. The hearing-impaired communication system depicted in Figure 1 is
`
`further illustrated in the block diagram of Figure 2. See Ex. 1003 at 2:53-54.
`
`
`
`Ex. 1003 at Figure 2. Figure 2 is a “block diagram of communication system 100.”
`
`Ex. 1003 at 4:44-45. It depicts the same relay service 110, hearing-impaired user’s
`
`device 120, and hearing-capable user’s device 190 as depicted in Figure 1. The
`
`
`
`-18-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 18 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`relay service “may include a display device 132 coupled to a processor-based
`
`device 102 comprising a processor 104 and a memory 106 . . . [and] an application
`
`program 108.” Ex. 1003 at 4:45-50. The hearing-impaired user’s device,
`
`“[c]ommunication device 120 may include a display device 134 and a processor-
`
`based device 152 comprising a processor 154 and a memory 156 . . . [and] an
`
`application program 118.” Ex. 1003 at 4:55-59. A person of ordinary skill in the
`
`art would recognize Figure 2 to disclose a single processor, 104, at the relay and a
`
`single processor, 154, at the hearing-impaired user’s device. In “operation of
`
`communication system 100, hearing-capable user 160 [] may convey speech into
`
`communication device 190, which may then transmit an associated voice signal to
`
`communication device 120 over network 180. Communication device 120 may
`
`then transmit the voice signal to relay service 110. Upon receipt of a voice signal,
`
`a human call assistant . . . may listen to the voice signal . . . and “revoice” the
`
`words to a speech recognition computer program . . . [which] may then output a
`
`text transcription . . . [which is] transmitted to communication device 120 via
`
`network 170.” Ex. 1003 at 4:64-5:13.
`
`37. As described above, the relay service 110 receives the voice signal of
`
`the hearing-capable user from the hearing-impaired user’s device 120 and
`
`generates a transcription of those words. Specifically, “upon receipt of a voice
`
`signal, a human call assistant [] positioned within relay service 110, may listen to
`
`
`
`-19-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 19 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`the voice signal transmitted from communication device 120 and “revoice” the
`
`words to a speech recognition computer program [] within relay service 110.” Ex.
`
`1003 at 5:2-7. During revoicing, “the call assistant repeats the words she or he
`
`hears upon receipt of the voice signal” and “[t]he speech recognition program . . .
`
`output[s] a text transcription of the call assistant’s spoken words.” Ex. 1003 at 5:8-
`
`11. This “text transcription may be transmitted to” the hearing-impaired user’s
`
`device for use by the hearing-impaired user during a conversation with the hearing-
`
`capable user. Ex. 1003 at 5:12-13.
`
`38. The ’689 Patent describes a “text transcription output” from a speech
`
`recognition program “as a ‘block’ of text.” Ex. 1003 at 5:14-15. Specifically, “a
`
`block of text may comprise one or more words, one or more sentences, one or
`
`more lines of text, or any combination thereof.” Ex. 1003 at 5:16-18. A person of
`
`ordinary skill in the art would understand that a speech recognition program would
`
`typically break the revoiced speech into blocks of limited size. Block size may be
`
`determined by a call assistant’s pauses during revoicing, the size of any provided
`
`transcription buffer, or the system preferences for providing the transcription to the
`
`hearing-impaired user with minimal delay. For example, a person of ordinary skill
`
`in the art could design a system that starts a new block each time the call assistant
`
`pauses for longer than a specified number of seconds. Alternatively, a system
`
`could be designed that starts new blocks at consistent time intervals, for example,
`
`
`
`-20-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 20 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`every five seconds during revoicing. Relatedly, the size of a block may be limited
`
`by the size of the device’s transcription buffer. Specifically, a block cannot be
`
`larger than the maximum size of the device’s transcription buffer because a
`
`transcription that exceeded that size would result in overflow and the loss of a
`
`portion of the transcription block. Similarly, blocks that are too large may impose
`
`excessive latency on the subsequent transmission of the text. This process will
`
`result in delay while the system waits for the full accumulation of the text
`
`comprising the block. Once the blocks of text have been generated by the speech
`
`recognition program they are “transmitted . . . [and] displayed within a text caption
`
`on display device 134,” at the hearing-impaired user’s device. Ex. 1003 at 5:23-
`
`25. The blocks “may also be displayed within a text caption displayed on display
`
`device 132,” at the relay. Ex. 1003 at 5:41-42.
`
`39. The ’689 Patent explains that the blocks of text may be displayed at
`
`the relay, and “any errors that may exist in the text caption displayed on display
`
`device 134 may also be visible to the” call assistant. Ex. 1003 at 5:45-47.
`
`Traditionally, there were two approaches to correcting errors within a text caption.
`
`A system could provide for correction by either (1) “backspacing an error in a text
`
`caption and displaying corrected text” or (2) “providing a corrected portion (e.g., a
`
`word or a sentence) at the end of a previously provided text caption.” Ex. 1003 at
`
`1:57-61. Each of these options presents challenges for the hearing-impaired user.
`
`
`
`-21-
`
`Sorenson IP Holdings Exhibit 2001
`Ultratec v. Sorenson IP Holdings IPR2017-01394
`Page 21 of 46
`
`
`
`IPR2017-01394
`U.S. Patent No. 9,336,689
`
`40. A person of ordinary skill in the art would understand the ’689
`
`Patent’s description of backspacing an error to be similar to corrections in a word
`
`processing program. Specifically, backspacing an error would involve selecting a
`
`specific portion of the text caption that includes an error, and using a backspace or
`
`delete key at the call assistant’s station to remove that error. While this method
`
`allows for correction of some types of errors, it also has some shortcomings. For
`
`example, the ’689 Patent explains that “backspacing an error in a text caption and
`
`displaying corrected text may provide a hearing-impaired user with a context for
`
`the correction . .