PTAB

IPR2025-00870

Samsung Electronics Co Ltd v. VB Assets LLC

Log In
Key Events
Petition
petition

1. Case Identification

2. Patent Overview

  • Title: Generating Natural Language System Responses Adapted Based on a User's Manner of Speaking
  • Brief Description: The ’699 patent discloses a cooperative conversational voice user interface that generates natural language responses. The system receives and recognizes a user’s spoken utterance, identifies context, determines an interpretation, and generates a response adapted to the user's manner of speaking based on accumulated short-term and long-term knowledge.

3. Grounds for Unpatentability

Ground 1: Obviousness over SmartKom and Kobsa - Claims 1-22 are obvious over SmartKom in view of Kobsa.

  • Prior Art Relied Upon: SmartKom (a 2006 book, SmartKom: Foundations of Multimodal Dialogue Systems) and Kobsa (a 1989 book, User Models in Dialog Systems).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner argued that SmartKom, a comprehensive overview of a multimodal dialogue system, disclosed the core elements of the challenged claims, including a system that receives user input, recognizes words, identifies context, determines an interpretation, and generates a response. SmartKom’s system used a “discourse state” (short-term knowledge) and a basic “user model” (long-term knowledge) to inform its interpretation and response. Petitioner contended that Kobsa, a survey of user modeling, provided the explicit teachings for robustly implementing the claimed long-term knowledge and user profile features that SmartKom only mentioned at a high level. Specifically, Kobsa taught incrementally building, storing, and updating long-term user models to supply other system components with assumptions about the user, thereby enabling tailored system behavior.
    • Motivation to Combine: A Person of Ordinary Skill in the Art (POSITA) would combine these references because SmartKom explicitly mentioned the utility of user models for context modeling and presentation planning but provided limited detail on their implementation. This would have motivated a POSITA to consult a foundational text like Kobsa to find detailed techniques for user modeling to improve SmartKom’s dialogue system. The goal was to enhance the user experience by providing more intelligent, user-specific, and tailored responses, as taught by Kobsa. The fact that a key editor was involved in both publications would have further directed a POSITA to the combination.
    • Expectation of Success: A POSITA would have had a reasonable expectation of success because SmartKom was built on a standard software architecture and Kobsa’s user models were described as software constructs. Integrating these known software techniques to improve a known type of system would have been a predictable endeavor.

Ground 2: Obviousness over Barbara, Ross, and Kellner - Claims 1-22 are obvious over Barbara in view of Ross and Kellner.

  • Prior Art Relied Upon: Barbara (Application # 2004/0101198), Ross (Application # 2002/0173960), and Kellner (Application # 2002/0065651).
  • Core Argument for this Ground:
    • Prior Art Mapping: Petitioner asserted that Barbara taught a voice interface system that interprets user intent from spoken commands and responds, meeting the basic claim elements. However, Barbara lacked architectural detail. Ross disclosed a robust architecture for a "conversation manager" that processes utterances, develops responses, and uses a "conversational record" to store dialogue history (short-term knowledge) that is eventually purged. Kellner taught a dialog system that adapts its output based on the user's "style of speech" (manner of speaking) by using an "associated user model" (long-term knowledge) to make the interaction more pleasant.
    • Motivation to Combine: A POSITA would combine Ross with Barbara to provide a concrete, flexible architecture for Barbara’s high-level system, particularly for managing conversational data and generating responses. This improved system would then be combined with Kellner's teachings to further enhance it. Kellner explicitly motivated adapting system outputs to a user's style to create a more pleasant experience and lower the "inhibition threshold" of system use. A POSITA would therefore have been motivated to incorporate Kellner’s user modeling and style-adaptation techniques into the Barbara/Ross system to make it more user-friendly and effective.
    • Expectation of Success: A POSITA would have had an expectation of success because the combination involved implementing known software techniques (Ross’s conversation manager, Kellner’s user model) into a known device (Barbara’s voice interface system) to achieve the predictable result of a more robust and user-adaptive system.

4. Key Claim Construction Positions

  • Petitioner argued that the term “speech recognition engine,” which appeared in dependent claims, should be construed consistent with a construction agreed upon in related district court litigation: “software or hardware that recognizes the words or phrases in the natural language utterance.” Petitioner stated no other terms required explicit construction.

5. Arguments Regarding Discretionary Denial

  • To address potential discretionary denial under Fintiv, Petitioner stipulated that if the Board institutes this IPR, Petitioner will not pursue in the related district court proceeding any ground that was raised or reasonably could have been raised against the challenged claims in the instituted IPR.

6. Relief Requested

  • Petitioner requests institution of an inter partes review and cancellation of claims 1-22 of Patent 10,755,699 as unpatentable under 35 U.S.C. §103.