DCT

2:25-cv-00621

VB Assets LLC v. Success Foods Management Group LLC

I. Executive Summary and Procedural Information

  • Parties & Counsel:
  • Case Identification: 2:25-cv-00621, E.D. Tex., 06/11/2025
  • Venue Allegations: Plaintiff alleges venue is proper in the Eastern District of Texas because Defendant has committed acts of infringement in the district and maintains a regular and established place of business there, citing a specific restaurant location in Tyler, Texas.
  • Core Dispute: Plaintiff alleges that Defendant’s voice-based artificial intelligence system for taking customer phone orders infringes four patents related to conversational voice interfaces and voice-based advertising.
  • Technical Context: The technology at issue involves natural language understanding (NLU) systems that can engage in cooperative dialogue, maintain context, and deliver targeted content like advertisements within a voice-based interaction.
  • Key Procedural History: The complaint details the history of Plaintiff’s predecessor, VoiceBox Technologies, as a pioneer in NLU, including its development of a prototype called "Cybermind" and its work with Toyota. Plaintiff also notes that the predecessor company was sold to Nuance Communications in 2018, but that ownership of the patents-in-suit was retained. No prior litigation or post-grant proceedings involving the asserted patents are mentioned in the complaint.

Case Timeline

Date Event
2006-10-16 Earliest Priority Date for ’341 Patent
2007-02-06 Earliest Priority Date for ’274, ’489, ’456 Patents
2012-03-27 ’489 Patent Issued
2013-09-03 ’274 Patent Issued
2019-12-17 ’341 Patent Issued
2024-12-05 Accused Torchy's Tacos Voice AI System Announced
2025-02-25 ’456 Patent Issued
2025-06-11 Complaint Filed

II. Technology and Patent(s)-in-Suit Analysis

U.S. Patent No. 10,510,341 - "SYSTEM AND METHOD FOR A COOPERATIVE CONVERSATIONAL VOICE USER INTERFACE"

The Invention Explained

  • Problem Addressed: Prior voice user interface systems were typically rigid "Command and Control" types that forced users to navigate verbal menus and memorize exact words or phrases to interact with the system, which was time-consuming and failed to provide a seamless conversational experience (Compl. ¶27; ’341 Patent, col. 1:61-2:15).
  • The Patented Solution: The invention describes a "cooperative conversational voice user interface" that can understand free-form human utterances. It does so by building upon and distinguishing between "short-term" knowledge (from the current conversation) and "long-term" knowledge (accumulated over time) to determine context, infer user intent, and generate adaptive, intelligent responses (Compl. ¶26, ¶30; ’341 Patent, col. 5:15-6:31).
  • Technical Importance: This approach aimed to move beyond simple command recognition toward a more natural, human-like dialogue model, making voice interfaces more intuitive and powerful (Compl. ¶31; ’341 Patent, col. 2:5-15).

Key Claims at a Glance

  • The complaint asserts at least independent Claim 10 (Compl. ¶26).
  • Essential elements of Claim 10 include:
    • accumulate short-term knowledge based on utterances received during a predetermined time period;
    • expire one or more items of short-term knowledge received prior to the predetermined time period;
    • accumulate long-term knowledge based on utterances received prior to the predetermined time period, where the long-term knowledge includes at least one expired item of short-term knowledge;
    • receive a first natural language utterance;
    • determine a first context for the utterance based on the short-term and long-term knowledge;
    • determine an interpretation of the utterance based on the context; and
    • generate a response based on the interpretation.
  • The complaint reserves the right to assert additional claims (Compl. ¶52).

U.S. Patent No. 8,527,274 - "SYSTEM AND METHOD FOR DELIVERING TARGETED ADVERTISEMENTS AND TRACKING ADVERTISEMENT INTERACTIONS IN VOICE RECOGNITION CONTEXTS"

The Invention Explained

  • Problem Addressed: The patent describes that prior voice interfaces were difficult to use, did not allow for cooperative dialogue when a user's needs were uncertain, and forced users to memorize specific commands. This created missed opportunities to provide users with valuable and relevant information, such as advertisements or purchase opportunities (’274 Patent, col. 1:50-67; Compl. ¶40, ¶41).
  • The Patented Solution: The invention provides a system that, upon receiving a natural language utterance containing an "incomplete or ambiguous" request, selects and presents one or more advertisements to the user. The system then monitors the user's interaction with the advertisement(s) and uses that interaction to interpret the original ambiguous request (Compl. ¶34; ’274 Patent, Abstract).
  • Technical Importance: This technology provides a method for monetizing voice interactions and resolving user ambiguity by presenting commercial options as a way to clarify intent, departing from menu-based systems toward context-based content selection (Compl. ¶43).

Key Claims at a Glance

  • The complaint asserts at least independent Claim 13 (Compl. ¶34).
  • Essential elements of Claim 13 include:
    • receive a natural language utterance from a voice-enabled device;
    • process the utterance to identify one or more requests;
    • determine that the utterance contains a request that is incomplete or ambiguous;
    • select one or more advertisements associated with the utterance;
    • output the selected advertisements to the voice-enabled device;
    • monitor interaction between a user and the advertisements; and
    • interpret the incomplete or ambiguous request based on the interaction.
  • The complaint reserves the right to assert additional claims (Compl. ¶57).

U.S. Patent No. 8,145,489 - "SYSTEM AND METHOD FOR SELECTING AND PRESENTING ADVERTISEMENTS BASED ON NATURAL LANGUAGE PROCESSING OF VOICE-BASED INPUT"

  • Technology Synopsis: This patent addresses selecting and presenting advertisements in response to a natural language utterance. It describes a system that interprets a voice request to establish a context and then delivers a relevant advertisement. The system then tracks user interaction with the ad to update cognitive models and build long-term and short-term shared knowledge for interpreting future utterances (Compl. ¶36).
  • Asserted Claims: The complaint asserts at least independent Claim 9 (Compl. ¶36).
  • Accused Features: The complaint accuses the Torchy's Tacos Voice AI Smart Ordering Systems of using conversational context to provide ordering suggestions, which Plaintiff frames as a form of advertisement or purchase opportunity (Compl. ¶44, ¶62).

U.S. Patent No. 12,236,456 - "SYSTEM AND METHOD FOR DELIVERING TARGETED ADVERTISEMENTS AND/OR PROVIDING NATURAL LANGUAGE PROCESSING BASED ON ADVERTISEMENTS"

  • Technology Synopsis: This patent describes a system for processing voice-based utterances to select and present purchase opportunities. The system determines a context from a first utterance, selects and delivers a purchase opportunity, tracks the user's interaction, and uses that interaction to update a user-specific profile. This profile is then used to interpret a subsequent utterance and select a subsequent purchase opportunity (Compl. ¶38).
  • Asserted Claims: The complaint asserts at least independent Claim 11 (Compl. ¶38).
  • Accused Features: The complaint accuses the entire Torchy's Tacos Voice AI Smart Ordering Systems, alleging it processes customer orders, presents purchase opportunities (e.g., menu suggestions), and uses interaction data to inform subsequent interactions (Compl. ¶44, ¶67).

III. The Accused Instrumentality

Product Identification

  • The "Torchy's Tacos Voice AI Smart Ordering Systems" ("Accused Systems") (Compl. ¶44).

Functionality and Market Context

  • The Accused Systems are described as a voice recognition and natural language understanding technology used by customers placing phone orders at all 130 Torchy's Tacos locations (Compl. ¶44, ¶45). The technology is implemented on servers that interact with mobile devices and is powered by SoundHound AI, Inc. (Compl. ¶45, ¶46).
    • A promotional screenshot illustrates the system engaging in a multi-turn dialogue to take a pickup order, including answering a user's question about popular dishes and suggesting a specific menu item (Compl., p. 14).
    • A system diagram shows that a "User Query" is processed via "Active Arbitration" and sent to both "Edge (Embedded)" and "SoundHound Cloud" components to generate a response (Compl., p. 15).
    • The complaint alleges the system was rolled out on December 5, 2024, and is marketed as an "easy-to-use option that ensures their orders are accurate, done on time" and improves the guest experience (Compl. ¶45, ¶50).

IV. Analysis of Infringement Allegations

U.S. Patent No. 10,510,341 Infringement Allegations

Claim Element (from Independent Claim 10) Alleged Infringing Functionality Complaint Citation Patent Citation
accumulate short-term knowledge based on one or more natural language utterances received during a predetermined time period The Accused Systems allegedly maintain context within a single phone order, such as remembering a user's request for a pickup order when they later ask about menu items. ¶44, ¶46 col. 5:17-21
expire one or more items of short-term knowledge that are based on one or more natural language utterances received prior to the predetermined time period The system's session-based knowledge from a single phone call is alleged to expire after the call ends, consistent with the claim. ¶26 col. 5:30-38
accumulate long-term knowledge...wherein the long-term knowledge includes at least one of the one or more expired items of short-term knowledge The Accused Systems allegedly build knowledge over multiple interactions (e.g., customer profiles, ordering history), incorporating data from past, expired sessions to inform future conversations. ¶26, ¶44 col. 5:39-49
receive a first natural language utterance via an input device The system receives voice utterances from customers over the phone via a mobile device. ¶44, ¶46 col. 2:34-36
determine a first context for the first natural language utterance based on the short-term knowledge and the long-term knowledge The system allegedly determines the context of a user's request (e.g., placing a food order) by using information from the current call and potentially from past interactions with that customer. ¶44, ¶46 col. 5:50-54
determine an interpretation...and generate a first response...based on the interpretation The system interprets user requests (e.g., "What's your most popular enchilada dish?") and generates a relevant response (e.g., "Our Enchiladas Supremas are a customer favorite!"). ¶46 col. 5:55-61
  • Identified Points of Contention:
    • Scope Questions: A central question may be whether the data handling in the Accused Systems, which are designed for transactional food ordering, meets the specific architectural requirements of "short-term knowledge," "long-term knowledge," and the "expiring" and "accumulating" steps as defined and enabled by the patent specification.
    • Technical Questions: The complaint alleges the accumulation of long-term knowledge on an information and belief basis. A key evidentiary question will be what proof exists that the Accused Systems actually create and utilize persistent, long-term user profiles that incorporate data from expired short-term sessions, as opposed to simply processing each call as a discrete, stateless transaction.

U.S. Patent No. 8,527,274 Infringement Allegations

Claim Element (from Independent Claim 13) Alleged Infringing Functionality Complaint Citation Patent Citation
receive a natural language utterance from a voice-enabled device The system receives customer voice commands and questions over the phone. ¶44, ¶46 col. 3:6-8
determine that the natural language utterance contains a request that is incomplete or ambiguous When a user asks a general question ("What's your most popular enchilada dish?") instead of placing a specific order, the system allegedly identifies this as an ambiguous or incomplete request. ¶46 col. 3:15-18
select one or more advertisements associated with the natural language utterance In response to the user's question, the system allegedly selects a specific menu item to recommend, which the plaintiff frames as a selected "advertisement" or purchase opportunity. ¶46 col. 3:19-21
output the one or more selected advertisements to the voice-enabled device The system speaks the recommendation to the user ("Our Enchiladas Supremas are a customer favorite!"). ¶46 col. 3:22-24
monitor interaction between a user and the one or more selected advertisements The system processes the user's subsequent response to the suggestion (e.g., "That sounds perfect, I'll take that."). ¶46 col. 3:25-27
interpret the request that is incomplete or ambiguous based on the interaction Based on the user's acceptance of the suggestion, the system interprets the original ambiguous request as a desire to order that specific item. ¶46 col. 3:28-30
  • Identified Points of Contention:
    • Scope Questions: The case may turn on whether a conversational suggestion about a standard menu item, offered by the direct seller of that item in response to a customer query, can be construed as an "advertisement" under the patent's claims. Defendant may argue this is merely part of the sales transaction itself, not a separate advertisement.
    • Technical Questions: What evidence does the complaint provide that a user question like "what's popular?" is technically processed as an "incomplete or ambiguous" request that triggers an ad selection module, as opposed to simply being a standard query handled by a general question-answering function?

V. Key Claim Terms for Construction

  • The Term: "short-term knowledge" and "long-term knowledge" (’341 Patent, Claim 10)
  • Context and Importance: The distinction between these two types of knowledge is foundational to the ’341 patent's claimed invention. The infringement theory depends on mapping the Accused Systems' data handling to this specific two-part architecture, including the claimed process of expiring short-term knowledge and selectively adding it to long-term knowledge. Practitioners may focus on whether these terms require distinct data stores and processes or can be read more broadly on any system that remembers information within a session versus across sessions.
  • Intrinsic Evidence for Interpretation:
    • Evidence for a Broader Interpretation: The claims define the terms functionally: short-term knowledge is "based on...utterances received during a predetermined time period," while long-term knowledge is based on utterances "received prior to the predetermined time period" (’341 Patent, col. 22:4-15). This could support a reading where any session-based data is "short-term" and any persistent data is "long-term."
    • Evidence for a Narrower Interpretation: The specification describes short-term knowledge accumulating "during a single conversation" and expiring after a "psychologically appropriate amount of time," while long-term knowledge is "user-centric, rather than session-based" and accumulates "over time" (’341 Patent, col. 5:15-43). This language may support a narrower construction requiring distinct session-based and user-centric data models.
  • The Term: "advertisement" (’274 Patent, Claim 13)
  • Context and Importance: The infringement theory for the ’274 patent and its asserted family members hinges on construing the Accused Systems' menu suggestions as "advertisements." If a conversational recommendation for a primary product by the direct seller is not an "advertisement," the infringement allegation may fail.
  • Intrinsic Evidence for Interpretation:
    • Evidence for a Broader Interpretation: The patent specification is part of a family that describes "purchase opportunities" (’456 Patent, Claim 11) and delivering commercial content. Plaintiff may argue that in the context of the invention, any communication presented to the user to induce a specific commercial transaction falls within the scope of an "advertisement."
    • Evidence for a Narrower Interpretation: The common understanding of "advertisement" often implies a message from a third party, or at least a distinct promotional message separate from the core product offering. Defendant may argue that a description of a menu item in response to a direct question is simply providing product information, not serving an advertisement as that term is understood in the art. The patent does not appear to provide an explicit definition of the term.

VI. Other Allegations

  • Indirect Infringement: The complaint alleges inducement of infringement for all four patents. It asserts that Torchy's Tacos actively and knowingly encourages and instructs its customers to use the Accused Systems in an infringing manner by advertising, marketing, and making the voice ordering systems available for placing phone orders (Compl. ¶53, ¶58, ¶63, ¶68).
  • Willful Infringement: The complaint does not contain an explicit count for willful infringement. However, it alleges that Defendant's inducement was done "actively and knowingly" (Compl. ¶53) and requests a declaration that the case is "exceptional" and an award of attorneys' fees under 35 U.S.C. § 285, which is often associated with findings of willful infringement or other litigation misconduct (Compl., Prayer for Relief ¶D).

VII. Analyst’s Conclusion: Key Questions for the Case

  1. A core issue will be one of definitional scope: Can a conversational menu suggestion made by a restaurant's AI in direct response to a customer's question be construed as an "advertisement" within the meaning of the ’274, ’489, and ’456 patents? The resolution of this claim construction dispute may be dispositive for a significant portion of the case.
  2. A second key issue will be one of architectural mapping: Does the Accused Systems' data management for phone orders practice the specific, multi-step process of accumulating, expiring, and transferring knowledge between distinct "short-term" and "long-term" stores as claimed by the ’341 patent, or is there a fundamental mismatch in technical operation?
  3. An evidentiary question will be one of proof of function: What evidence will be produced to demonstrate that the Accused Systems create and utilize persistent, cross-session "long-term knowledge" profiles for users, as alleged in the complaint, versus treating each phone order as a largely independent, session-based transaction?