Amazon.com Inc v. VB Assets LLC

1. Case Identification

• Case #: IPR2020-01367
• Patent #: 8,073,681
• Filed: July 28, 2020
• Petitioner(s): Amazon.com, Inc.; Amazon.com LLC; Amazon Web Services, Inc.; A2Z DEVELOPMENT CENTER, INC. D/B/A LAB126; Rawles LLC; AMZN Mobile LLC; AMZN Mobile 2 LLC; AMAZON.COM SERVICES, INC. F/K/A AMAZON FULFILLMENT SERVICES, INC.; AND AMAZON.COM SERVICES LLC (formerly AMAZON DIGITAL SERVICES LLC)
• Patent Owner(s): VB Assets, LLC
• Challenged Claims: 1-42

2. Patent Overview

• Title: SYSTEM AND METHOD FOR A COOPERATIVE CONVERSATIONAL VOICE USER INTERFACE
• Brief Description: The ’681 patent discloses a human-to-machine voice user interface that uses both short-term knowledge (from the current conversation) and long-term knowledge (from user profiles) to understand and respond to user utterances within an identified context.

3. Grounds for Unpatentability

Ground 1: Obviousness over Kennewick - Claims 1-2, 9, 13-14, 21, 25-26, 33, 37, 39, and 41 are obvious over Kennewick.

• Prior Art Relied Upon: Kennewick (Application # 2004/0193420).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Kennewick, which describes a speech-based natural language query environment, teaches all limitations of the independent claims. Kennewick allegedly discloses accumulating short-term knowledge via a "dialog history" for the current conversation and accumulating long-term knowledge via extensive "user profile" data that includes past interactions. Petitioner asserted Kennewick uses both knowledge sources to determine a context for a user's utterance (e.g., distinguishing "temperature" as weather vs. a measurement) and to disambiguate intent. Kennewick then generates a natural language response adapted to this determined context, for instance by asking clarifying questions.
  • Motivation to Combine (for §103 grounds): This ground is asserted under 35 U.S.C. §103 based on a single reference, arguing that Kennewick alone renders the claims obvious.
  • Expectation of Success (for §103 grounds): Petitioner asserted that a Person of Ordinary Skill in the Art (POSITA) would have understood that implementing the disclosed features of Kennewick for a conversational interface was straightforward.

Ground 2: Obviousness over Kennewick and Huang - Claims 5, 8, 17, 20, 29, and 32 are obvious over Kennewick in view of Huang.

• Prior Art Relied Upon: Kennewick (Application # 2004/0193420) and Huang ("Spoken Language Processing: A Guide to Theory, Algorithm and System Development," Prentice Hall (2001)).
• Core Argument for this Ground:
  • Prior Art Mapping: Petitioner argued that Huang supplements Kennewick by teaching a "plan-based" methodology for discourse analysis. This involves identifying a "conversational goal," the roles of participants (e.g., system-directed vs. user-directed), and the allocation of information between them. Huang allegedly teaches classifying an utterance into a "conversation type" based on these factors to better establish the user's intended meaning. Petitioner asserted these teachings map directly to the limitations of dependent claim 5.
  • Motivation to Combine: A POSITA would combine Kennewick with Huang to improve the inferential capabilities of Kennewick's system. Huang's plan-based approach provides a more robust contextual tool for understanding user intent, aligning with Kennewick's goal of creating a natural and flexible user interaction by minimizing the required user input.
  • Expectation of Success: A POSITA would have had a reasonable expectation of success in incorporating Huang's well-known discourse analysis techniques into Kennewick’s conversational framework to enhance its performance.

Ground 3: Obviousness over Kennewick and Seneff - Claims 10, 22, 34, 38, 40, and 42 are obvious over Kennewick in view of Seneff.

• Prior Art Relied Upon: Kennewick (Application # 2004/0193420) and Seneff ("Hypothesis Selection and Resolution in the Mercury Flight Reservation System," MIT (2001)).

• Core Argument for this Ground:

  • Prior Art Mapping: Petitioner argued Seneff adds a specific hypothesis selection and verification process to Kennewick. Seneff's system generates multiple preliminary interpretations of an utterance ("N-best candidate frames") with associated confidence scores. It then selects an initial interpretation with the highest confidence level. If this interpretation is found to be incorrect (e.g., through user confirmation), it is removed from the short-term knowledge (dialog history), and the system determines the intended meaning based on the interpretation with the next-highest confidence level.
  • Motivation to Combine: A POSITA would combine Seneff's hypothesis scoring system with Kennewick to improve the accuracy of understanding user utterances. While Kennewick teaches using context to resolve ambiguity, Seneff provides a specific, well-understood method for systematically evaluating and selecting from multiple potential meanings, directly furthering Kennewick’s goal of providing reliable and expected results.
  • Expectation of Success: The combination was allegedly predictable, as Kennewick’s context-rich system would provide the necessary information (dialog history, user profile) to support the evaluation and confidence-scoring of hypotheses as taught by Seneff.

• Additional Grounds: Petitioner asserted additional obviousness challenges, including: claims 6-7, 18-19, and 30-31 over Kennewick, Huang, and Seneff; claims 11-12, 23-24, and 35-36 over Kennewick in view of Coffman (Patent 6,839,896) for adding multi-modal inputs/outputs; and claims 3-4, 15-16, and 27-28 over Kennewick in view of Mitsuyoshi (Application # 2003/0182123) for expiring short-term information from a context stack over time.

4. Key Claim Construction Positions

• "implicit hypothesis" (claims 38, 40, 42): Petitioner argued this term should be construed as "a potential intent of a user utterance generated by filling in missing information using contextual data." This construction is based on the ’681 patent’s disclosure that implicit hypotheses are generated when an utterance is missing information needed to complete a task, which the system then infers from short-term or long-term knowledge.

5. Arguments Regarding Discretionary Denial

• Petitioner argued against discretionary denial under §314(a) or §325(d). It contended that denying institution would undermine AIA objectives because the petition challenges claims 1-42, whereas the related district court litigation only alleges infringement of claim 41.
• Regarding §325(d), Petitioner argued that although Kennewick was cited in an Information Disclosure Statement (IDS) during prosecution, it was one of over 370 references disclosed and was never substantively considered by the examiner. Therefore, the Becton factors weigh against discretionary denial.

6. Relief Requested

• Petitioner requests institution of an inter partes review and cancellation of claims 1-42 of the ’681 patent as unpatentable.