PTAB

IPR2020-00578

Nice Ltd v. CallMiner Inc

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Analyzing Spoken Words
  • Brief Description: The ’509 patent describes a system for analyzing verbal communications by using speech recognition to generate multiple "word alternatives" and associated probabilities for spoken words. The patent asserts that its novelty lies in storing this data in a database for subsequent analysis, such as identifying word frequencies, whereas prior systems allegedly discarded this information.

3. Grounds for Unpatentability

Ground 1: Obviousness over Viswanathan, Baker, and Sato - Claims 1, 6-8, 13, and 14

  • Prior Art Relied Upon: Viswanathan (Patent 6,345,253), Baker (Patent 5,680,511), and Sato (Patent 6,212,517).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that the combination of these references taught all elements of independent claims 1 and 8. Viswanathan taught a system that transcribes audio files, identifies "best" and "alternative" word choices, assigns them "confidence values," and stores them in permanent index files for later retrieval and analysis. Baker disclosed a speech recognition system that generates a list of "choice words" with corresponding "probability signals" for each spoken word and stores them in permanent memory. Sato described text retrieval methods based on computing word frequencies within single documents and across an entire database. Petitioner contended that these references collectively disclosed producing an electronic recording, processing it to identify word alternatives with probabilities, loading this data to a database, and examining the data to determine characteristics like word frequency, as required by the independent claims.
    • Motivation to Combine: A POSITA would combine Viswanathan and Baker because they addressed the same problem using conventional and overlapping speech recognition techniques. Baker provided a concrete implementation of a speech recognizer that Viswanathan assumed as a component of its system. A POSITA would have been motivated to look to Sato’s text-based retrieval system to learn about and implement the background indexing and retrieval processes that Viswanathan’s audio-based system sought to emulate. The combination represented a predictable substitution of known elements for their intended purposes.
    • Expectation of Success: A POSITA would have had a high expectation of success because combining the references involved applying known techniques to achieve predictable results. For example, using Baker's explicit probability signals in Viswanathan’s system, which already used analogous "confidence values," would have been a straightforward implementation.

Ground 2: Obviousness over Viswanathan, Baker, Sato, and Young - Claims 2 and 9

  • Prior Art Relied Upon: Viswanathan (Patent 6,345,253), Baker (Patent 5,680,511), Sato (Patent 6,212,517), and Young (WO 01/74042).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground added Young to the base combination to teach the dependent claim limitation of comparing word frequencies "in a plurality of time periods." Young disclosed analyzing transcribed telephone calls from a call center and specifically taught using a "time-of-day feature" to segment calls into different time periods (e.g., daytime vs. nighttime) to enable comparison of term frequencies across those periods. Young also described tracking clusters of calls over time to detect emerging trends.
    • Motivation to Combine: A POSITA would have been motivated to incorporate Young’s temporal analysis into the base system of Viswanathan/Baker/Sato to gain the important ability to identify trends in spoken communications over time. Since Viswanathan and Sato already taught updating statistics for new documents, adding a time parameter as taught by Young would have been a predictable and straightforward enhancement.

Ground 3: Obviousness over Viswanathan, Baker, Sato, and Cragun - Claims 3-5 and 10-12

  • Prior Art Relied Upon: Viswanathan (Patent 6,345,253), Baker (Patent 5,680,511), Sato (Patent 6,212,517), and Cragun (Patent 6,970,863).
  • Core Argument for this Ground:
    • Prior Art Mapping: This ground added Cragun to the base combination to teach the limitation of detecting the "presence of a pattern, the pattern comprising a defined sequence of word alternatives." Cragun explicitly disclosed a document retrieval system that enabled users to construct searches with Boolean operators, including the use of quotation marks to search for an exact phrase. This functionality directly corresponded to searching for a "defined sequence of word alternatives."
    • Motivation to Combine: A POSITA would combine Cragun's functionality with the base references to provide a more robust and conventional search capability. Since Baker and Sato already taught Boolean keyword searches, extending this to include well-known phrase searching (as taught by Cragun) would have been an obvious and desirable improvement to meet user expectations. The petition further argued that implementing this common feature into the existing frameworks would have been predictable.

4. Arguments Regarding Discretionary Denial

  • Petitioner argued that discretionary denial would be inappropriate because the proposed prior art grounds were not considered during the original prosecution. It was asserted that none of the primary references (Viswanathan, Baker, Sato, Cragun) were applied to the challenged claims by the Examiner. While Young was cited in an Information Disclosure Statement (IDS), Petitioner contended it was neither discussed nor applied against any claim, a circumstance that weighs heavily in favor of institution.

5. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 1-14 of Patent 7,346,509 as unpatentable.