1:23-cv-00052
Voxware Inc v. Honeywell Intl Inc
I. Executive Summary and Procedural Information
- Parties & Counsel:- Plaintiff: Voxware, Inc. (Delaware)
- Defendant: Honeywell International Inc., Hand Held Products, Inc., Intermec Inc., and Vocollect, Inc. (Delaware and Pennsylvania)
- Plaintiff’s Counsel: Irwin IP LLP; Richards, Layton & Finger, P.A.
 
- Case Identification: 1:23-cv-00052, D. Del., 04/26/2023
- Venue Allegations: Plaintiff alleges venue is proper in the District of Delaware because the Defendant entities are incorporated in Delaware, reside there, and/or have agents in the district.
- Core Dispute: Plaintiff seeks a declaratory judgment that its voice-directed warehouse logistics products do not infringe, and that the claims are invalid and unenforceable, for five U.S. patents owned by Defendants, while also asserting claims for antitrust violations, deceptive trade practices, and tortious interference.
- Technical Context: The dispute centers on voice-directed warehouse management systems, where workers use headsets to receive spoken instructions and confirm task completion, a technology critical for efficiency and accuracy in the logistics and supply chain industry.
- Key Procedural History: The complaint details a multi-year history of pre-suit correspondence, initiated by Defendants in February 2020, alleging infringement of the patents-in-suit. Plaintiff alleges that these infringement allegations, culminating in a letter dated December 1, 2022, directly caused the cancellation of a planned acquisition of Plaintiff by a third party.
Case Timeline
| Date | Event | 
|---|---|
| 2001-12-31 | Alleged earliest date of public use for Voxware's VoiceLogistics and VISE software | 
| 2005-02-04 | Earliest Priority Date for ’219 Patent | 
| 2005-02-14 | Earliest Priority Date for ’669 Patent | 
| 2006-02-06 | Earliest Priority Date for ’419 Patent | 
| 2009-10-27 | ’669 Patent Issued | 
| 2011-02-08 | ’419 Patent Issued | 
| 2011-05-20 | Earliest Priority Date for ’290 Patent | 
| 2012-08-28 | ’219 Patent Issued | 
| 2012-12-31 | Alleged earliest date of public use for Voxware's VoxPilot/VoxConsole software | 
| 2014-10-15 | Earliest Priority Date for ’490 Patent | 
| 2014-12-16 | ’290 Patent Issued | 
| 2020-02-13 | Defendants send "First Honeywell Letter" alleging infringement by VISE software | 
| 2020-03-13 | Plaintiff sends "First Voxware Response" denying infringement | 
| 2021-02-02 | ’490 Patent Issued | 
| 2021-12-22 | Defendants send "Second Honeywell Letter" re-asserting claims and adding allegations re: ’490 Patent | 
| 2022-01-26 | Plaintiff sends "Second Voxware Response" denying infringement | 
| 2022-12-01 | Defendants send "Third Honeywell Letter" changing accused products for certain patents | 
| 2022-12-31 | Planned acquisition of Plaintiff is allegedly cancelled due to Third Honeywell Letter | 
| 2023-04-26 | Plaintiff files First Amended Complaint for Declaratory Judgment | 
II. Technology and Patent(s)-in-Suit Analysis
U.S. Patent No. 8,255,219 - "Method and Apparatus for Determining a Corrective Action for a Speech Recognition System Based on the Performance of the System" (Issued Aug. 28, 2012)
The Invention Explained
- Problem Addressed: The patent addresses the need to assess and improve the performance of speech recognition systems, which can be affected by factors like background noise or a user's unfamiliarity with the system (Compl. ¶113; ’219 Patent, col. 1:40-44).
- The Patented Solution: The invention proposes a method where a processor determines the performance of a speech recognition system by analyzing its success in recognizing specific words or sets of words. Based on this performance assessment, the system then determines a "corrective action" to improve future performance (’219 Patent, Abstract). The system can generate a performance report and suggest actions like retraining a specific word (’219 Patent, col. 9:10-48; Fig. 4).
- Technical Importance: This technology aims to automate the optimization of speech recognition systems, potentially reducing the need for manual intervention by supervisors and improving user efficiency in voice-directed work environments (Compl. ¶113).
Key Claims at a Glance
- Independent Claim 1 is asserted (Compl. ¶37, ¶66-67). Its essential elements are:- A method for improving performance of a speech recognition system comprising:
- operating a processor to determine a performance of the system based on either recognition of instances of a word or recognition of instances of various words among a set of words, and
- determining a corrective action based on the performance, to improve the performance.
 
U.S. Patent No. 8,914,290 - "Systems and Methods for Dynamically Improving User Intelligibility of Synthesized Speech in a Work Environment" (Issued Dec. 16, 2014)
The Invention Explained
- Problem Addressed: Synthesized speech from a text-to-speech (TTS) engine can be difficult for users to understand, particularly in noisy industrial environments or when the user is unfamiliar with a prompt (’290 Patent, col. 1:40-54).
- The Patented Solution: The invention describes a system that monitors at least one environmental condition related to the intelligibility of the TTS output (e.g., ambient noise, user commands like "say again," or prompt type). In response, the system automatically modifies an operational parameter of the TTS engine, such as its volume or speed, to make the speech easier to understand, and can later restore the original setting (’290 Patent, Abstract; col. 4:5-24).
- Technical Importance: This approach seeks to enhance user productivity and reduce errors by making voice-directed systems adaptive to changing real-world conditions without requiring constant manual adjustment by the user (Compl. ¶75).
Key Claims at a Glance
- Independent Claim 12 is asserted (Compl. ¶37, ¶74-75). Its essential elements are:- A method of communicating in a speech-based environment using a text-to-speech engine, the method comprising:
- monitoring at least one environmental condition associated with a user that is related to intelligibility of an audible output of the text-to-speech engine by the user; and
- modifying at least one adjustable operational parameter of the text-to-speech engine in response to the monitored at least one environmental condition to improve the intelligibility of an audible output of the text-to-speech engine.
 
U.S. Patent No. 10,909,490 - "Systems and Methods for Worker Resource Management" (Issued Feb. 2, 2021)
- Technology Synopsis: The patent describes a system for managing worker resources by recording user activity from a voice-directed terminal, including break durations and idle times. This data is analyzed to identify productivity patterns, predict work completion times, and implement resource decisions, such as reassigning workers to different tasks or regions (’490 Patent, Abstract; col. 3:5-20).
- Asserted Claims: Independent claim 17 is asserted (Compl. ¶46).
- Accused Features: Defendants allegedly accused Voxware’s Intellestra and VoxPilot software, which Plaintiff contends were discontinued or in public use prior to the patent's issuance and do not perform the claimed step of implementing worker resource decisions (Compl. ¶46, ¶48, ¶82, ¶84).
U.S. Patent No. 7,609,669 - "Voice Directed System and Method Configured for Assured Messaging to Multiple Recipients" (Issued Oct. 27, 2009)
- Technology Synopsis: The technology concerns a method for delivering a message in a speech-enabled environment. The core of the invention is a closed-loop confirmation process where a terminal outputs an audio message, receives a spoken verbal confirmation from the user indicating the message was understood, and transmits an acknowledgement message reflecting that confirmation (’669 Patent, Abstract; col. 10:11-20).
- Asserted Claims: Independent claim 19 is asserted (Compl. ¶55).
- Accused Features: Defendants allegedly accused Voxware’s VoxConnect and VMS software. Plaintiff alleges its products do not have the required features of receiving a message, receiving a verbal confirmation, and transmitting an acknowledgement based on that confirmation (Compl. ¶55, ¶93).
U.S. Patent No. 7,885,419 - "Headset Terminal with Speech Functionality" (Issued Feb. 8, 2011)
- Technology Synopsis: This patent describes a specific mechanical assembly for a headset. The invention focuses on a rotatable microphone boom that couples with the earcup assembly via a retainer that includes a "plurality of snaps" which capture a "flange" on the boom assembly, allowing it to snap together and rotate (’419 Patent, Abstract; col. 18:8-21).
- Asserted Claims: Independent claim 8 is asserted (Compl. ¶57).
- Accused Features: Defendants allegedly accused Voxware’s BTH430 headset. Plaintiff alleges the BTH430 does not have the claimed "plurality of snaps" or a boom assembly with a "flange that is captured by the snaps" (Compl. ¶101-102).
III. The Accused Instrumentality
Product Identification
The complaint identifies several accused instrumentalities, reflecting an evolving set of allegations from Defendants over time (Compl. ¶37, ¶46, ¶55, ¶57, ¶63). The products include:
- Software: VISE (Voxware Integrated Speech Engine), Voice Management Suite (VMS), Intellestra, VoxPilot, and VoxConnect.
- Hardware: BTH430 headset.
Functionality and Market Context
- The accused software products (VISE, VMS, VoxPilot, VoxConnect, Intellestra) provide voice-directed solutions for managing and monitoring operations in distribution centers and warehouses (Compl. ¶4). These products direct workers via voice commands through headsets to perform tasks like order picking.
- The complaint alleges that these software products are part of a competitive "Voice Logistics Market" and that key functionalities, such as those in VISE and VMS, were sold or in public use by at least December 2001, predating the asserted patents (Compl. ¶11, ¶44, ¶59).
- The BTH430 is a headset product used in conjunction with this software. The complaint provides photographic evidence to illustrate its mechanical construction (Compl. ¶102; p. 34, Fig. 1). Figure 1 of the complaint shows a side view of the BTH430's boom assembly, intended to demonstrate the mechanism by which it connects to the headset housing (Compl. p. 34).
IV. Analysis of Infringement Allegations
U.S. Patent No. 8,255,219 Infringement Allegations
| Claim Element (from Independent Claim 1) | Alleged Infringing Functionality | Complaint Citation | Patent Citation | 
|---|---|---|---|
| operating a processor to determine a performance of the system based on either recognition of instances of a word or recognition of instances of various words among a set of words | Plaintiff alleges that Defendants mapped this element to the VISE software's "training experience" and noise-cancelling functions. | ¶40 | col. 4:45-55 | 
| and determining a corrective action based on the performance, to improve the performance. | Plaintiff alleges that Defendants mapped this element to the VISE software's "training experience" and noise-cancelling functions. | ¶40 | col. 4:56-59 | 
Identified Points of Contention
- Scope Questions: The primary dispute raises the question of whether the VISE software's general "training experience" constitutes "determin[ing] a performance... based on... recognition of... words" as required by the claim (Compl. ¶40, ¶67). A court may need to determine if a preliminary user setup process meets the claim's requirement for an ongoing performance evaluation.
- Technical Questions: Plaintiff’s allegations suggest a technical mismatch, questioning what evidence shows that the accused software's functionalities perform the specific two-step process of first evaluating word recognition performance and then, based on that specific evaluation, determining a corrective action (Compl. ¶67).
U.S. Patent No. 8,914,290 Infringement Allegations
| Claim Element (from Independent Claim 12) | Alleged Infringing Functionality | Complaint Citation | Patent Citation | 
|---|---|---|---|
| monitoring at least one environmental condition associated with a user that is related to intelligibility of an audible output of the text-to-speech engine by the user | Plaintiff alleges Defendants pointed to the VISE software's noise-cancelling functions and manual user volume controls. | ¶41 | col. 7:4-11 | 
| and modifying at least one adjustable operational parameter of the text-to-speech engine in response to the monitored at least one environmental condition... | Plaintiff alleges Defendants pointed to the user's ability to manually input voice commands to adjust the headset's volume. | ¶41 | col. 7:12-20 | 
Identified Points of Contention
- Scope Questions: The analysis may turn on whether a user manually adjusting volume qualifies as the system "modifying" an operational parameter "in response to" a "monitored" condition (Compl. ¶41, ¶75). This raises a claim construction question about the degree of automation required by the claim language.
- Technical Questions: Plaintiff’s position raises the factual question of whether the accused VISE software automatically monitors for conditions related to intelligibility (like ambient noise levels or user difficulty) and then autonomously adjusts TTS parameters, or if it relies solely on manual user intervention (Compl. ¶75).
V. Key Claim Terms for Construction
For the ’219 Patent
- The Term: "determining a corrective action based on the performance"
- Context and Importance: The viability of the infringement allegation hinges on whether the accused "training experience" or "noise cancelling function" meets this limitation (Compl. ¶40). Practitioners may focus on this term because it requires a causal link between a measured "performance" and a subsequent "corrective action," a link Plaintiff alleges is absent in its products (Compl. ¶67).
- Intrinsic Evidence for Interpretation:- Evidence for a Broader Interpretation: The specification states that corrective actions can be "suggestions to a user" or can be "automatically initiated" (’219 Patent, col. 4:56-62), which could be argued to encompass a wide range of system responses.
- Evidence for a Narrower Interpretation: The patent’s examples of corrective actions include specific, data-driven instructions such as "Retrain Word 5" based on a calculated performance grade (’219 Patent, Fig. 4). This may support an interpretation requiring a more specific and analytical process than a general training setup.
 
For the ’290 Patent
- The Term: "modifying... in response to the monitored... condition"
- Context and Importance: This term is critical because Plaintiff alleges its software does not automatically modify parameters based on monitoring, but rather allows for manual user adjustments (Compl. ¶41, ¶75). The dispute may center on whether the claim requires system autonomy.
- Intrinsic Evidence for Interpretation:- Evidence for a Broader Interpretation: The patent states the modification can be based on "user input such as voice commands" (’290 Patent, col. 4:21-22), which might be argued to cover scenarios where a user's command (e.g., "say again") is the "monitored condition" that triggers a modification (e.g., replaying the prompt slower).
- Evidence for a Narrower Interpretation: The detailed description repeatedly discusses the system detecting an "adverse environment" (e.g., a noisy freezer) and then automatically increasing volume or changing language settings, suggesting a system-driven, not user-driven, process (’290 Patent, col. 9:44-67; col. 12:45-54).
 
VI. Other Allegations
Indirect Infringement
The complaint states Defendants alleged that Voxware’s VISE software "at least contributes to or induces infringement" of the ’219 and ’290 Patents and that its Intellestra product does the same for the ’490 Patent (Compl. ¶37, ¶46). Voxware’s complaint seeks a declaratory judgment of non-infringement for direct and indirect infringement (Compl. ¶68, ¶76, ¶86).
Willful Infringement
Willfulness is not alleged by Defendants in the pre-suit letters cited in the complaint. However, Plaintiff’s complaint alleges that Defendants' assertions of infringement were made without a good faith basis and were "objectively baseless," forming part of Plaintiff's claims for antitrust violations and tortious interference (Compl. ¶112-117).
VII. Analyst’s Conclusion: Key Questions for the Case
- A central issue will be one of temporal scope and invalidity: does the functionality of the accused Voxware software suites (VISE, VMS, VoxPilot) predate the 2005 and later priority dates of the asserted patents, as Plaintiff alleges? If proven, this could invalidate the asserted claims under 35 U.S.C. § 102 based on prior public use.
- A key question of claim scope will be whether the functions of the accused software meet the specific, multi-step processes required by the asserted method claims. For the ’219 and ’290 patents, this will likely involve determining whether general software features like user training and manual volume control can be construed to meet limitations requiring automated performance assessment and responsive parameter modification.
- A dispositive factual question will be one of structural identity: does the accused BTH430 headset contain the specific "plurality of snaps" capturing a "flange" as recited in claim 8 of the ’419 patent? The photographic evidence included in the complaint suggests this will be a direct comparison of the physical product against the claim language.