Jawbone Innovations LLC v. ZTE Corp

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: Jawbone Innovations, LLC (Texas)
  • Defendant: ZTE Corporation (China)
  • Plaintiff’s Counsel: Fabricant LLP; McKool Smith, P.C.
• Case Identification: 2:23-cv-00082, E.D. Tex., 02/28/2023
• Venue Allegations: Plaintiff alleges venue is proper because Defendant is a foreign company that may be sued in any judicial district, and further alleges that Defendant has committed acts of patent infringement and maintains sufficient minimum contacts within the Eastern District of Texas.
• Core Dispute: Plaintiff alleges that Defendant’s earbuds and smartphones infringe eight patents related to acoustic noise suppression and voice activity detection technologies.
• Technical Context: The technology at issue addresses methods for improving audio clarity in communication devices by algorithmically separating a user's speech from ambient background noise, a key feature in the consumer and professional audio markets.
• Key Procedural History: The complaint alleges the patented technology was originally developed by AliphCom (later Jawbone, Inc.) for military applications to enable clear communication in combat conditions. Following Jawbone, Inc.'s liquidation in 2017, Plaintiff alleges that Defendant was contacted regarding its infringement of the patent portfolio and the portfolio's availability for purchase or license. These allegations of pre-suit notice may form the basis for claims of willful infringement.

Case Timeline

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

U.S. Patent No. 8,019,091 - "Voice Activity Detector (VAD)-Based Multiple-Microphone Acoustic Noise Suppression"

• Issued: September 13, 2011.

The Invention Explained

• Problem Addressed: Conventional multi-microphone noise suppression systems struggle to distinguish between desired speech and unwanted noise, particularly when the noise characteristics change. (Compl. ¶20; ’091 Patent, col. 1:11-29).
• The Patented Solution: The invention uses a sensor that detects human tissue vibration to create a highly accurate Voice Activity Detector (VAD). This VAD provides a definitive signal indicating when a user is speaking. The system leverages this information to generate a "first transfer function" when only noise is present and a "second transfer function" when speech is present. By using these context-specific functions, the system can more effectively remove noise from the acoustic signal. (Compl. ¶20; ’091 Patent, Abstract; col. 4:21-35).
• Technical Importance: This approach provided a robust method for noise cancellation in demanding environments, such as for military communications, where traditional acoustic-only methods could fail. (Compl. ¶15).

Key Claims at a Glance

• The complaint asserts independent claim 11. (Compl. ¶35).
• Essential elements of claim 11 include:
  • A receiver for at least two acoustic signals from at least two acoustic microphones.
  • At least one sensor that receives human tissue vibration information associated with voicing activity.
  • A processor that generates a plurality of transfer functions, including:
    • A first transfer function (based on a ratio of energy between microphones) generated when voicing is absent.
    • A second transfer function generated when voicing is present.
  • Removing acoustic noise using the first transfer function and a combination of the first and second transfer functions to produce a denoised data stream.

U.S. Patent No. 7,246,058 - "Detecting Voiced and Unvoiced Speech Using Both Acoustic and Nonacoustic Sensors"

• Issued: July 17, 2007.

The Invention Explained

• Problem Addressed: Traditional systems have difficulty distinguishing not only between speech and noise, but also between "voiced" speech (which involves vocal cord vibration) and "unvoiced" speech (fricatives like 's' or 'f' sounds, which do not). (Compl. ¶52; ’058 Patent, col. 1:22-34).
• The Patented Solution: The invention uses two distinct methods for classifying sounds. It employs a non-acoustic "voicing sensor" to detect physiological vibration; by cross-correlating this sensor's data with the acoustic signal from a microphone, it identifies voiced speech. Concurrently, it analyzes the difference in signal gain between two microphones to distinguish unvoiced speech from ambient noise. (Compl. ¶¶19, 52; ’058 Patent, Abstract; col. 1:37-56).
• Technical Importance: By providing a reliable way to differentiate between voiced speech, unvoiced speech, and noise, the system enables more sophisticated and effective noise suppression algorithms. (Compl. ¶19).

Key Claims at a Glance

• The complaint asserts independent claim 1. (Compl. ¶52).
• Essential elements of claim 1 include:
  • At least two microphones receiving acoustic signals.
  • At least one voicing sensor receiving physiological information.
  • At least one processor that performs three distinct identification steps:
    • Identifies voiced speech by cross-correlating the physiological information and an acoustic signal and exceeding a correlation threshold.
    • Identifies unvoiced speech when a difference parameter (based on signal gain between microphones) exceeds a gain threshold.
    • Identifies noise when the difference parameter is less than the gain threshold.

U.S. Patent No. 10,779,080 - "Dual Omnidirectional Microphone Array (DOMA)"

• Issued: September 15, 2020.
• Technology Synopsis: This patent describes using an array of two physical omnidirectional microphones to create two "virtual" microphones. These virtual microphones are designed to have substantially similar responses to noise but substantially dissimilar responses to speech. This difference allows an adaptive noise removal application to more effectively isolate and remove noise from the user's speech. (Compl. ¶22).
• Asserted Claims: The complaint asserts at least claim 1. (Compl. ¶70).
• Accused Features: The complaint alleges that ZTE products, such as the LiveBuds Pro, use two physical omnidirectional microphones and a processor to generate two beamformed virtual microphones that infringe the patent. (Compl. ¶¶23, 71-76).

U.S. Patent No. 11,122,357 - "Forming Virtual Microphone Arrays Using Dual Omnidirectional Microphone Array (DOMA)"

• Issued: September 14, 2021.
• Technology Synopsis: This patent relates to creating distinct virtual directional microphones from the signals of two physical microphones. The virtual microphones are configured to have similar responses to noise but dissimilar responses to speech. A signal processor then combines signals from the virtual microphones by filtering and summing them in the time domain to generate an output signal with attenuated noise. (Compl. ¶24).
• Asserted Claims: The complaint asserts at least claim 1. (Compl. ¶85).
• Accused Features: The complaint alleges that ZTE products use arrays of physical microphones to form beamformed virtual microphones that have similar noise responses and dissimilar speech responses, and use a processor to combine their signals to reduce noise. (Compl. ¶¶25, 86-89).

U.S. Patent No. 8,467,543 - "Microphone and Voice Activity Detection (VAD) Configurations for Use with Communications Systems"

• Issued: June 18, 2013.
• Technology Synopsis: This patent describes a communications system with a voice detection subsystem and a denoising subsystem. The denoising subsystem uses a microphone array where microphones are fixed at different positions and orientations relative to a user's mouth. The voice detection subsystem uses a sensor (independent of the microphone array) to generate control signals that tell the denoising subsystem when to subtract a noise estimate from the acoustic signal. (Compl. ¶26).
• Asserted Claims: The complaint asserts at least claim 1. (Compl. ¶98).
• Accused Features: The complaint alleges ZTE products utilize a speech-detecting accelerometer to control a denoising system that uses a microphone array with at least one microphone oriented toward the user's mouth and another oriented away. (Compl. ¶¶27, 99-105).

U.S. Patent No. 8,503,691 - "Virtual Microphone Arrays Using Dual Omnidirectional Microphone Array (DOMA)"

• Issued: August 6, 2013.
• Technology Synopsis: This patent describes forming two virtual microphones from two physical omnidirectional microphones. The first virtual microphone has a linear response to both speech and noise. The second virtual microphone is designed to have a "null" oriented toward the speech source, making it insensitive to speech, while maintaining a linear response to noise similar to the first virtual microphone. This dissimilarity in speech response allows for effective noise cancellation. (Compl. ¶24).
• Asserted Claims: The complaint asserts at least claim 23. (Compl. ¶115).
• Accused Features: The complaint alleges ZTE products create beamformed microphones from physical omnidirectional microphones, where one has a linear response to speech and the other has a null directed at the user's mouth. (Compl. ¶¶25, 116-121).

U.S. Patent No. 8,321,213 - "Acoustic Voice Activity Detection (AVAD) for Electronic Systems"

• Issued: November 27, 2012.
• Technology Synopsis: This patent describes an acoustic voice activity detection system that uses two physical microphones to form two virtual microphones. A filter is created based on calibration and delay between the microphone signals. This filter is then used to form a second virtual microphone. The system determines that voice activity is present when an energy ratio between the first and second virtual microphones exceeds a threshold. (Compl. ¶28).
• Asserted Claims: The complaint asserts at least claim 1. (Compl. ¶131).
• Accused Features: The complaint alleges ZTE products use filter-and-sum beamforming to create virtual microphones and detect user speech by comparing a ratio of energies from these microphones to a threshold, for example, to detect a wake word. (Compl. ¶¶29, 132-135).

U.S. Patent No. 8,326,611 - "Acoustic Voice Activity Detection (AVAD) for Electronic Systems"

• Issued: December 4, 2012.
• Technology Synopsis: This patent describes a method for acoustic voice activity detection that largely mirrors the system described in the ’213 Patent. It involves forming a first virtual microphone from two physical microphones, forming a filter describing the relationship between them, using that filter to form a second virtual microphone, generating an energy ratio between the two virtual microphones, and detecting voice activity when that ratio exceeds a threshold. (Compl. ¶28).
• Asserted Claims: The complaint asserts at least claim 1. (Compl. ¶145).
• Accused Features: The complaint alleges ZTE products perform the claimed method by using beamformed microphones and detecting voice activity (like a wake word) when the energy ratio between them surpasses a threshold. (Compl. ¶¶29, 146-149).

III. The Accused Instrumentality

Product Identification

• The complaint identifies "at least all versions and variants of ZTE earbuds (e.g., LiveBuds Pro, LiveBuds, etc.) and smartphones (e.g., Axon 40 Ultra, Axon 40 Pro, Axon 30, Axon 30 5G, etc.)" as the Accused Products. (Compl. ¶31). The ZTE LiveBuds Pro is used as the primary exemplary product throughout the infringement allegations.

Functionality and Market Context

• The accused functionality is the noise suppression and voice activity detection features in these products. (Compl. ¶31). The complaint alleges that the ZTE LiveBuds Pro, for example, incorporates an array of microphones, an accelerometer, and a voice pickup sensor to implement these features. (Compl. ¶21). The complaint provides a marketing image for the ZTE LiveBuds Pro describing it as a "True Hybrid Three Microphone Noise Reduction Headset" capable of bringing "Clear Call." (Compl. p. 11). The alleged technical operation involves using the microphones in a beamforming array to generate virtual microphones and using the accelerometer to detect when a user is speaking, thereby enabling the system to distinguish speech from noise. (Compl. ¶¶21, 27, 29).

IV. Analysis of Infringement Allegations

U.S. Patent No. 8,019,091 Infringement Allegations

• Identified Points of Contention:
  • Scope Questions: A central question may be the construction of the term "transfer function." The infringement theory depends on whether the accused products' noise suppression algorithm can be characterized as generating two distinct transfer functions—one for noise-only conditions and another for speech-plus-noise conditions—as specified by the claim.
  • Technical Questions: What evidence does the complaint provide that the accused processor's operation maps directly onto the claimed method? The complaint alleges the system applies different functions based on the VAD state (Compl. ¶¶40-42), but the actual implementation of the noise removal algorithm (e.g., "least mean squares method" alleged in ¶42) will be scrutinized to determine if it meets the specific claim limitations regarding the use and combination of these functions.

U.S. Patent No. 7,246,058 Infringement Allegations

• Identified Points of Contention:
  • Scope Questions: The construction of "voicing sensor" will be a focal point. The dispute may turn on whether a general-purpose accelerometer in a consumer earbud, as alleged in the complaint, meets the claim's requirement of a sensor that "receives physiological information associated with human voicing activity."
  • Technical Questions: The claim requires a processor that performs a three-part classification scheme: voiced speech, unvoiced speech, and noise, using two different inputs (cross-correlation and gain difference). A key technical question will be whether the accused products' processor actually executes this specific logic, or if it uses a different algorithm to process signals from its various sensors to achieve noise reduction.

V. Key Claim Terms for Construction

For the ’091 Patent

• The Term: "transfer function"
• Context and Importance: This term is the core of the invention's logic. Claim 11 requires the processor to generate a "first transfer function" when voicing is absent and a "second transfer function" when voicing is present. The outcome of the case may depend on whether the accused product's algorithm can be shown to generate and use two distinct, context-dependent mathematical representations of the acoustic environment that meet the definition of "transfer function."
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The specification describes the transfer function generally as representing the relationship between acoustic signals at different microphones. (’091 Patent, col. 4:11-20). This could support an argument that any algorithm that models this relationship meets the limitation.
  • Evidence for a Narrower Interpretation: The detailed description repeatedly links the generation of the first transfer function to periods when "voicing activity is absent" and the second to when it is "present." (’091 Patent, col. 4:21-30). This could support a narrower construction requiring two explicitly different and separately generated algorithmic states or models tied directly to the VAD signal.

For the ’058 Patent

• The Term: "voicing sensor that receives physiological information"
• Context and Importance: This term distinguishes the invention from purely acoustic-based systems. The infringement case hinges on whether the accused accelerometer and/or "voice pickup sensor" (Compl. ¶38) falls within this definition. Practitioners may focus on whether "physiological information" requires direct measurement of biological processes (like vocal cord vibration) or if it can broadly cover any vibration detected on the user's body.
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The patent abstract refers to "non-acoustic sensors that receive physiological information," suggesting the key distinction is non-acoustic, which could encompass any sensor detecting body-conducted vibrations like an accelerometer. (’058 Patent, Abstract).
  • Evidence for a Narrower Interpretation: The background discusses the shortcomings of purely "acoustic" analysis. (’058 Patent, col. 1:22-34). This context could be used to argue that the "voicing sensor" must be something fundamentally different from a microphone, and its "physiological information" must be more specific than just any vibration that an accelerometer might pick up, such as from head movement.

VI. Other Allegations

• Indirect Infringement: The complaint alleges that ZTE induces infringement by providing the Accused Products to customers and distributors along with "instruction manuals, websites, promotional materials, advertisements, and other information" that demonstrate how to use the allegedly infringing noise cancellation features. (Compl. ¶¶44, 62). It is alleged that ZTE is aware that the normal and customary use of the products would infringe the patents. (Compl. ¶¶44, 62).
• Willful Infringement: The complaint alleges willful infringement based on knowledge of the patents "since at least 2017 but by no later than the date of this Complaint." (Compl. ¶¶44, 62). This knowledge is alleged to have arisen from communications with ZTE following the liquidation of Jawbone, Inc., which included notification of ZTE's infringement and the availability of the portfolio for license or purchase. (Compl. ¶¶18, 46, 64).

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

• A key evidentiary question will be one of algorithmic functionality: Do the accused ZTE products operate by generating distinct, VAD-gated "transfer functions" as required by the '091 Patent, and do they perform the specific three-part sound classification (voiced, unvoiced, noise) via two different physical inputs as required by the '058 Patent? Or do they achieve a similar noise-cancelling result through a technically different process that falls outside the claims' scope?
• A central issue of claim construction will be one of definitional scope: Can the term "voicing sensor," rooted in the patent's context of capturing "physiological information," be construed to cover the general-purpose accelerometers and voice pickup sensors found in the accused consumer earbuds and smartphones?
• A crucial factual question for willfulness and damages will be the nature of pre-suit notice: What was the specific content of the alleged communications to ZTE beginning in 2017, and did these communications provide a sufficiently particularized notice of infringement to support a finding of willful misconduct for any subsequent infringement?