DCT

1:22-cv-01623

CDN Innovations LLC v. SoundHound Inc

Log In

I. Executive Summary and Procedural Information

  • Parties & Counsel:
  • Case Identification: 1:22-cv-01623, D. Del., 12/22/2022
  • Venue Allegations: Venue in the District of Delaware is alleged on the grounds that Defendant is a Delaware corporation and has committed acts of infringement and maintains a regular and established place of business in the judicial district, though the complaint cites a Georgia address for that place of business.
  • Core Dispute: Plaintiff alleges that Defendant’s voice assistant and music recognition technologies infringe five patents related to dynamically formatting information for display devices, recognizing spoken identifiers using predefined grammars, and identifying and delivering music based on ambient audio samples.
  • Technical Context: The technologies at issue are foundational to modern voice-enabled artificial intelligence and music identification services, a market where automated, user-friendly interaction is critical.
  • Key Procedural History: The complaint alleges that Defendant had pre-suit knowledge of the patents-in-suit via letters sent in March and April 2020, which identified Plaintiff's patents and, in one instance, specifically alleged infringement of U.S. Patent No. 7,870,088.

Case Timeline

Date Event
2000-01-31 Earliest Priority Date for ’088, ’397, and ’208 Patents
2000-03-28 Priority Date for ’180 Patent
2001-09-19 Priority Date for ’532 Patent
2001-10-30 U.S. Patent No. 6,311,180 Issues
2005-03-08 U.S. Patent No. 6,865,532 Issues
2011-01-11 U.S. Patent No. 7,870,088 Issues
2013-08-13 U.S. Patent No. 8,509,397 Issues
2016-01-01 Alleged Infringement Begins (on or about this date)
2019-04-30 U.S. Patent No. 10,275,208 Issues
2020-03-20 Plaintiff allegedly sends first notice letter to Defendant
2020-04-06 Plaintiff allegedly sends second notice letter to Defendant
2022-12-22 Complaint Filed

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

U.S. Patent No. 6,311,180 - Method for mapping and formatting information for a display device, Issued October 30, 2001

The Invention Explained

  • Problem Addressed: The patent describes the "conformal problem" arising from the proliferation of communication devices (e.g., cell phones, PDAs) with varied display sizes and capabilities, which makes it difficult to display a single source document effectively across all devices (Compl. ¶10; ’180 Patent, col. 1:29-51).
  • The Patented Solution: The invention proposes a method to dynamically generate a "display document" that conforms to a specific device's limitations and a user's viewing preferences (’180 Patent, Abstract). Instead of creating multiple static versions of a document, the system uses a generalized "application description file" and a mapping system to select and arrange content based on the target device's geometry, fonts, and the user's preferred language or character size (’180 Patent, col. 3:26-34; col. 4:46-54).
  • Technical Importance: This approach provided a framework for content delivery to the nascent mobile internet, where adapting information to small, non-standardized screens was a significant technical and commercial challenge (’180 Patent, col. 1:52-65).

Key Claims at a Glance

  • The complaint asserts independent claim 1 (Compl. ¶19).
  • Essential elements of claim 1 (method):
    • Dynamically creating a display document to fit a display device based on the device's limitations and user's viewing preferences.
    • Providing source contents in a predetermined format.
    • Recognizing the display limitations from a first information source.
    • Determining the user's viewing preferences from a second information source.
    • Selecting preferred display contents from the source contents using a mapping system.
    • Generating the final display document containing the selected contents.
  • The complaint reserves the right to assert additional claims (Compl. ¶18).

U.S. Patent No. 6,865,532 - Method for recognizing spoken identifiers having predefined grammars, Issued March 8, 2005

The Invention Explained

  • Problem Addressed: The patent addresses the difficulty for voice-operated devices to correctly recognize long sequences of spoken numbers, such as telephone or account numbers, and the need for a more robust method for voice-based device control (Compl. ¶12; ’532 Patent, col. 2:40-44).
  • The Patented Solution: The invention discloses a method for creating and recognizing unique spoken identifiers by defining a phrase with a predetermined grammatical structure composed of multiple "word slots" (e.g., number, adjective, noun) (’532 Patent, Abstract). Each slot is associated with a set of unique words (e.g., "three," "white," "horse"), and a unique identifier is formed by concatenating one word from each set, creating a memorable and error-resistant phrase like "three white horse swam around Chicago" (’532 Patent, col. 5:1-13).
  • Technical Importance: This structured, grammar-based approach to identifiers offered a more reliable alternative to digit-string recognition, potentially improving the usability of voice control systems in an era before advanced natural language processing was widespread (’532 Patent, col. 5:36-42).

Key Claims at a Glance

  • The complaint asserts independent claim 1 (Compl. ¶28).
  • Essential elements of claim 1 (method):
    • Defining a phrase having a plurality of word slots arranged in a predetermined order and according to a predetermined grammatical structure.
    • Associating a set of unique words with each word slot according to the grammatical structure.
    • Generating a plurality of unique identifiers by selecting one word from each set for each word slot.
  • The complaint reserves the right to assert additional claims (Compl. ¶27).

Multi-Patent Capsule: U.S. Patent No. 7,870,088 - Method of Delivering Music and Information, Issued January 11, 2011

  • Technology Synopsis: This patent, along with the related ’397 and ’208 patents, addresses the problem of identifying and obtaining music a user hears ambiently (e.g., on the radio, in a store) without knowing the title or artist (Compl. ¶14; ’088 Patent, col. 1:21-32). The solution involves a portable device that samples the music, transmits the sample to a server for identification against a database, and then delivers the identified music and related information (e.g., concert dates, merchandise) to a user-specified device (’088 Patent, col. 1:33-41).
  • Asserted Claims: Independent claims 1 and 64 (Compl. ¶40).
  • Accused Features: Defendant's "Accused Music Delivery Technologies" are alleged to infringe (Compl. ¶39).

Multi-Patent Capsule: U.S. Patent No. 8,509,397 - Apparatus and Methods of Delivering Music and Information, Issued August 13, 2013

  • Technology Synopsis: Sharing a specification with the ’088 patent, this patent covers a system for identifying ambient music. A portable device captures an audio sample, transmits it to a host computer that searches a database to identify the music, and the host computer then transmits the identified music to one or more reception units (’397 Patent, Abstract).
  • Asserted Claims: Claims 13 and 24 (Compl. ¶51).
  • Accused Features: Defendant's "Accused Music Delivery Technologies" are alleged to infringe (Compl. ¶51).

Multi-Patent Capsule: U.S. Patent No. 10,275,208 - Apparatus and Methods of Delivering Music and Information, Issued April 30, 2019

  • Technology Synopsis: Also in the same family as the ’088 and ’397 patents, this patent describes a method where a computing system receives a speech command, increases microphone sensitivity in response, captures non-speech audio (music), and matches it to an identified audio pattern (’208 Patent, cl. 1). This patent focuses more specifically on the voice-activated workflow for initiating a music search.
  • Asserted Claims: Independent claim 1 (Compl. ¶64).
  • Accused Features: Defendant's "Accused Music Delivery Technologies" are alleged to infringe (Compl. ¶63).

III. The Accused Instrumentality

Product Identification

  • The complaint identifies the accused instrumentalities categorically as Defendant's "Accused Display Instrumentalities," "Accused Speech Recognition Instrumentalities," and "Accused Music Delivery Technologies" (Compl. ¶¶18, 27, 39). These categories appear to encompass Defendant's software and services that provide voice-enabled assistance and music identification.

Functionality and Market Context

  • The complaint alleges that the accused products and services are sold and used throughout the United States (Compl. ¶4, ¶18, ¶27). The accused functionalities include generating display documents that conform to display devices, recognizing spoken identifiers, and delivering music and related information based on audio samples (Compl. ¶¶10, 12, 14). The complaint alleges infringing use has occurred since at least 2016 (Compl. ¶22, ¶34, ¶46). No probative visual evidence provided in complaint.

IV. Analysis of Infringement Allegations

The complaint references preliminary infringement charts in Exhibits A-2 and B-2, which were not provided with the filed document. The infringement theory is therefore summarized based on the complaint's narrative allegations.

U.S. Patent No. 6,311,180 Infringement Allegations

The complaint alleges that Defendant's use of the "Accused Display Instrumentalities" infringes at least claim 1 of the ’180 Patent (Compl. ¶¶18, 20). The narrative theory suggests that when Defendant's services provide information to a user's device, they dynamically format that information to fit the specific device's display, thereby practicing the patented method (Compl. ¶10).

  • Identified Points of Contention:
    • Scope Questions: A central question may be whether the modern process of a server sending data via an API to a native mobile application, which then renders the user interface, constitutes "dynamically creating a display document" as understood in the context of the patent's 2000 priority date. The defense may argue that the "document" is not created by the accused system in the manner claimed, but rather rendered by the user's own device software.
    • Technical Questions: Claim 1 requires using a "first information source" for display limitations and a "second information source" for user preferences. A factual dispute may arise over whether Defendant's system actually segregates these inputs or uses a more integrated method that does not map to the claim's distinct steps.

U.S. Patent No. 6,865,532 Infringement Allegations

The complaint alleges that Defendant's "Accused Speech Recognition Instrumentalities" infringe at least claim 1 of the ’532 Patent (Compl. ¶¶27, 29). The infringement theory is that Defendant's voice-controlled services practice the claimed method of recognizing spoken identifiers that have predefined grammars (Compl. ¶12).

  • Identified Points of Contention:
    • Scope Questions: The core dispute may focus on the term "phrase having a plurality of word slots." The patent describes a highly structured, template-like system (e.g., number-adjective-noun). The question will be whether this language can be construed to cover modern natural language understanding (NLU) systems, which typically use probabilistic models and flexible grammars rather than rigid, concatenative "word slots."
    • Technical Questions: What evidence does the complaint provide that Defendant's system generates identifiers by "selecting one word from each set for each word slot"? This suggests a combinatorial generation process, which may be technically distinct from how NLU systems parse and interpret the semantic intent of unstructured user speech.

V. Key Claim Terms for Construction

Term: "display document" ('180 Patent, claim 1)

  • Context and Importance: The definition of this term is critical for determining whether the patent applies to modern app-based interfaces or is limited to the web-based document models (e.g., WML, HTML) prevalent at the time of invention. The infringement analysis for the ’180 patent hinges on whether the data and instructions sent to SoundHound's app constitute a "display document."
  • Intrinsic Evidence for Interpretation:
    • Evidence for a Broader Interpretation: The specification states the mapping system generates a document "in a computer language that is supported by the display device" and provides WML as one example, suggesting the term is not limited to a single format (’180 Patent, col. 6:16-20). The term could be argued to encompass any data structure whose purpose is to cause information to be displayed.
    • Evidence for a Narrower Interpretation: The patent repeatedly uses "web pages" as its primary example (’180 Patent, col. 2:5-10) and discusses creating a "single web page that is formatted to display on a target display device" (’180 Patent, col. 5:10-12). This context may support a narrower construction limited to self-contained, navigable documents rather than components of an integrated application interface.

Term: "predetermined grammatical structure" ('532 Patent, claim 1)

  • Context and Importance: This term is central to defining the scope of the claimed invention. Practitioners may focus on this term because its construction will determine whether the claim reads on flexible NLU systems or is confined to the rigid, rule-based grammar exemplified in the patent.
  • Intrinsic Evidence for Interpretation:
    • Evidence for a Broader Interpretation: The claim language itself is general and does not specify the type of grammatical structure. An argument could be made that any system that parses speech into semantic components (e.g., intent, entities) is using a form of "grammatical structure."
    • Evidence for a Narrower Interpretation: The specification provides a specific example of the structure: "number:adjective:noun:verb:preposition:proper-noun" (’532 Patent, col. 5:1-2). This explicit, ordered structure could be used to argue that the claim requires a fixed, template-based grammar, not the fluid grammars processed by modern AI assistants.

VI. Other Allegations

Indirect Infringement

  • The complaint alleges both induced and contributory infringement for the ’532, ’088, ’397, and ’208 patents. Inducement is based on allegations that Defendant encourages use of its technologies through advertising and technical documentation (Compl. ¶31, ¶43, ¶55, ¶67). Contributory infringement is based on allegations that Defendant's software constitutes a material component of the invention that is not a staple article of commerce suitable for substantial non-infringing use (Compl. ¶32, ¶44, ¶56, ¶68).

Willful Infringement

  • The complaint alleges that Defendant had knowledge of the asserted patent families as early as March and April 2020 from notice letters sent by Plaintiff (Compl. ¶30, ¶42, ¶54, ¶66). These allegations of pre-suit knowledge establish a factual basis for a potential claim of willful infringement and for the request that the case be declared "exceptional" under 35 U.S.C. § 285 (Compl. p. 15).

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

  • A core issue will be one of definitional scope: can claim terms rooted in the technological context of the early 2000s—such as "display document" (’180 patent) and a "phrase having a plurality of word slots" (’532 patent)—be construed to cover the functionally different paradigms of modern mobile applications and probabilistic natural language processing?
  • A second central question will be one of technological translation: for the music identification patents (’088, ’397, ’208), the case will likely turn on whether the plaintiff can demonstrate that SoundHound’s complex audio fingerprinting and matching technology practices the specific sequence of sampling, transmitting a signal derived from ambient music, and server-side matching as recited in the claims.
  • A potential threshold issue may concern pleading sufficiency and precision, particularly regarding the complaint's inconsistent venue allegations, which could invite early motion practice and judicial scrutiny of the complaint's drafting.