Nice Ltd v. CallMiner Inc

1. Case Identification

• Case #: IPR2020-00616
• Patent #: 8,583,434
• Filed: February 24, 2020
• Petitioner(s): Nice Ltd.
• Patent Owner(s): CallMiner, Inc.
• Challenged Claims: 1-28

2. Patent Overview

• Title: Analyzing Verbal Communication
• Brief Description: The ’434 patent describes systems for analyzing recorded conversations. The technology uses speech recognition to identify multiple "word alternatives" and an associated "probability" for each spoken word, then stores this data in an electronic file for subsequent analysis, such as word counting.

3. Grounds for Unpatentability

Ground 1: Obviousness over Core Speech Recognition References - Claims 1, 12, and 21 are obvious under 35 U.S.C. §103 over Viswanathan, Baker, and Bowater.

• Prior Art Relied Upon: Viswanathan (Patent 6,345,253), Baker (Patent 5,680,511), and Bowater (Patent 6,278,772).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that the combination of these references taught all limitations of the independent claims. Viswanathan was asserted to teach an audio retrieval system that transcribes audio files, stores the "best" and "alternative" word choices in index files with associated "confidence" values, and analyzes this data for retrieval. Baker was cited for its speech recognition system that generates a list of "choice words" with corresponding "probability signals" for each spoken word and stores them in memory. Bowater was used to supply the context of recording and transcribing telephone conversations between customers and call center agents, fulfilling the "plurality of conversations" element.
  • Motivation to Combine: A POSITA would combine Viswanathan and Baker as they were prominent, commercially-known systems (from IBM and Dragon Systems, respectively) that used conventional techniques to solve the same problem of speech recognition and retrieval, making them readily interchangeable or combinable. A POSITA would have been motivated to apply this combined system in the call center environment disclosed by Bowater (another IBM patent) to produce and analyze the type of audio files described in Viswanathan.
  • Expectation of Success: Combining these known components for their intended purposes—using Baker's speech recognition to generate inputs for Viswanathan's audio indexing system in Bowater's call center context—would have yielded only predictable results.

Ground 2: Obviousness with Added Text Retrieval Methods - Claims 2, 3, 13, 14, 22, and 23 are obvious over Viswanathan, Baker, Bowater, and Sato.

• Prior Art Relied Upon: Viswanathan (Patent 6,345,253), Baker (Patent 5,680,511), Bowater (Patent 6,278,772), and Sato (Patent 6,212,517).
• Core Argument for this Ground:
  • Prior Art Mapping: This ground built upon the combination in Ground 1, adding Sato to teach the dependent claim limitations related to determining and comparing word frequencies. Petitioner asserted that Sato discloses maintaining a "global statistic table" to count word frequencies across all documents and, in response to a query, calculating frequencies within a set of retrieved texts to determine each word's "importance" and rank them.
  • Motivation to Combine: A POSITA would combine Sato with the primary references because Viswanathan's retrieval system emulated the two-phase indexing and retrieval approach taught by Sato. It would have been obvious to look to Sato to implement or understand the underlying techniques for frequency analysis that Viswanathan used for document scoring and ranking.
  • Expectation of Success: The combination was predictable as both Viswanathan and Sato use substantially the same statistical features (e.g., term frequency, inverse document frequency) for information retrieval, making the integration of Sato's specific frequency counting methods into Viswanathan's system straightforward.

Ground 3: Obviousness with Added Temporal Analysis - Claims 4, 15, and 24 are obvious over Viswanathan, Baker, Bowater, Sato, and Young.

• Prior Art Relied Upon: The references from Ground 2, plus Young (WO 01/74042).

• Core Argument for this Ground:

  • Prior Art Mapping: This ground added Young to the prior combination to teach the limitation of comparing word frequencies "in a plurality of time periods." Petitioner argued that Young discloses transcribing telephone calls and using a "time-of-day feature" to divide calls into "workday" and "night" buckets, enabling the system to compare word frequencies across these different time periods to identify trends.
  • Motivation to Combine: A POSITA would be motivated to add Young's temporal analysis to the existing system to track term frequencies over time—a known concern in the art for identifying trends. Young addressed the same problems as Viswanathan and recommended speech recognition systems from Dragon Systems, the common assignee of Baker, making it a natural source for a POSITA to consult.
  • Expectation of Success: Adding a time parameter to the existing statistical data tracked by Viswanathan and Sato would have been a simple modification for a POSITA, leading to the predictable result of enabling trend analysis over time.

• Additional Grounds: Petitioner asserted an additional obviousness challenge (Ground 4) for claims 5-11, 16-20, and 25-28 based on the combination of Viswanathan, Baker, Bowater, and Cragun (Patent 6,970,863). This ground relied on Cragun for its teachings on user-defined query weighting and graphical user interfaces for search.

4. Arguments Regarding Discretionary Denial

• Petitioner argued that the Board should not exercise discretionary denial under 35 U.S.C. §325(d) because the asserted grounds were not cumulative to art considered during prosecution. Petitioner emphasized that the core prior art references—Viswanathan, Baker, Bowater, and Cragun—were never presented to or considered by the Examiner.
• While the Young reference was cited during prosecution, Petitioner contended that it was being applied in a new context for a narrow teaching in dependent claims, not for the independent claims as the Examiner had applied it. Furthermore, the independent claims were significantly amended during prosecution to overcome the rejection based on Young, making the current challenge distinct.

5. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 1-28 of the ’434 patent as unpatentable.