PTAB
IPR2020-01346
Amazon.com Inc v. VB Assets LLC
1. Case Identification
- Patent #: 9,015,049
- Filed: July 23, 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: 1-20
2. Patent Overview
- Title: SYSTEM AND METHOD FOR A COOPERATIVE CONVERSATIONAL VOICE USER INTERFACE
- Brief Description: The ’049 patent discloses a conversational voice user interface that processes natural language utterances. The system uses "short-term knowledge" accumulated during a single conversation and "long-term knowledge" from prior user interactions to determine context, interpret the user's intent, and generate an appropriate response.
3. Grounds for Unpatentability
Ground 1: Obviousness over Kennewick - Claims 1-6, 10-16, and 20 are obvious over Kennewick, alone or in view of the knowledge of a POSITA.
- Prior Art Relied Upon: Kennewick (Application # 2004/0193420).
- Core Argument for this Ground:
- Prior Art Mapping: Petitioner argued that Kennewick, which was filed by the same original applicant as the ’049 patent, discloses all limitations of the challenged claims. Kennewick describes a conversational speech system that receives natural language queries and provides responses. Petitioner contended that Kennewick’s "dialog history," which is built during a conversation and used for context, is equivalent to the ’049 patent’s "short-term knowledge" in a "first model." Further, Kennewick's use of this dialog history to determine context, interpret a query (e.g., distinguishing "temperature" as weather vs. measurement), and select a "domain agent" to formulate a response directly maps to the core steps of independent claims 1 and 11. For claims requiring "long-term knowledge" (e.g., claim 4), Petitioner mapped this to Kennewick's disclosure of creating, storing, and using an extensive "user profile" for each user, which includes a history of questions, session histories, and interactions with the system over time.
- Motivation to Combine (for §103 grounds): As this ground primarily relies on a single reference, the motivation was presented as Kennewick itself teaching all the claimed features. The argument for obviousness over Kennewick alone was framed as an alternative to anticipation, asserting that even if not every limitation is explicitly disclosed, it would have been obvious for a person of ordinary skill in the art (POSITA) to implement the claimed system based on Kennewick's detailed disclosures.
- Expectation of Success (for §103 grounds): A POSITA would have had a high expectation of success because Kennewick provides a comprehensive blueprint for a conversational system, and implementing its features as claimed would involve applying known techniques for their intended purposes.
Ground 2: Obviousness over Kennewick and Cooper - Claims 7-9 and 17-19 are obvious over Kennewick in view of Cooper.
- Prior Art Relied Upon: Kennewick (Application # 2004/0193420) and Cooper (Patent 6,757,362).
- Core Argument for this Ground:
- Prior Art Mapping: This ground addresses dependent claims that add the limitation of identifying the "manner" in which an utterance is spoken and using that manner to generate a response. Petitioner argued that while Kennewick provides the foundational conversational system (as detailed in Ground 1), Cooper teaches the specific missing element: adapting system behavior based on the user's "manner." Cooper discloses a virtual assistant that identifies the user's manner by detecting politeness (e.g., "please," "thank you") or emotional state (from voice volume, word choice, speech rate). Cooper’s system then adapts its response to match, for example, by using polite discourse or including submissive words like "sorry" if the user is angry. Petitioner asserted that combining Cooper's manner-detection and adaptive response features with Kennewick's system would render claims 7-9 and 17-19 obvious.
- Motivation to Combine (for §103 grounds): A POSITA would combine these references to improve the user experience of the Kennewick system. Kennewick explicitly states a goal of making responses more "natural" and simulating aspects of human "personality." Cooper directly addresses this by teaching how to make a virtual assistant's responses more appropriate and adaptive based on the user's manner of speaking. Combining Cooper's teachings was presented as a predictable solution to achieve Kennewick's stated goal.
- Expectation of Success (for §103 grounds): A POSITA would have a reasonable expectation of success in integrating Cooper's manner-analysis module into Kennewick's system, as both references operate in the same field of conversational interfaces and the combination represents the application of a known improvement (manner adaptation) to a known system.
4. Arguments Regarding Discretionary Denial
- Petitioner argued that the Board should not exercise its discretion to deny institution under 35 U.S.C. §314 or §325. The core argument was that although the primary reference, Kennewick, was cited in an Information Disclosure Statement (IDS) during the original prosecution of the ’049 patent, it was never substantively considered by the Examiner. Petitioner highlighted that Kennewick was listed among more than 600 references identified during prosecution and was included in an IDS spanning nearly 25 pages. The Examiner's Office Actions never discussed Kennewick, and therefore, the reference's teachings were not before the USPTO in the way they are presented in the petition.
5. Relief Requested
- Petitioner requested institution of an inter partes review and cancellation of claims 1-20 of Patent 9,015,049 as unpatentable.