Promptu Systems Corp v. Comcast Corp

I. Executive Summary and Procedural Information

• Parties & Counsel:
  • Plaintiff: Promptu Systems Corporation (Delaware)
  • Defendant: Comcast Corporation (Pennsylvania) and Comcast Cable Communications, LLC (Delaware)
  • Plaintiff’s Counsel: White and Williams LLP; Carr / Ferrell, LLP
• Case Identification: 2:16-cv-06516, E.D. Pa., 12/19/2016
• Venue Allegations: Plaintiff alleges venue is proper in the Eastern District of Pennsylvania because Defendants maintain a principal place of business in the district, conduct continuous and systematic business, provide the accused infringing products and services to residents of the district, and maintain offices and employees there.
• Core Dispute: Plaintiff alleges that Defendant’s XFINITY and X1 Entertainment Operating Systems infringe three patents related to voice recognition technology for controlling television services.
• Technical Context: The technology concerns architectures for implementing voice command features in cable television systems, a significant user interface component in the modern interactive media market.
• Key Procedural History: The complaint alleges a detailed pre-suit history between the parties, beginning around 2001, involving a confidential relationship, technical disclosures by Plaintiff, and multiple written agreements. Plaintiff alleges that Defendant terminated this relationship around 2006 before launching a "substantially similar" voice recognition system in 2015, which may be central to the allegations of willful infringement.

Case Timeline

Date	Event
2000-06-08	Priority Date for ’196 and ’326 Patents
~2001	Comcast allegedly first expressed interest in Promptu’s technology
2002-01-08	Priority Date for ’538 Patent
2003-05-06	Memorandum of Understanding allegedly signed by Comcast and Promptu
2004-02-09	Addendum to Memorandum of Understanding allegedly signed
2005-05-11	Marketing Trial and License and Development Agreements allegedly executed
2006-05-16	U.S. Patent No. 7,047,196 Issues
~2006	Comcast allegedly terminated voice recognition implementation by Promptu
2007-08-21	U.S. Patent No. 7,260,538 Issues
2013-06-25	U.S. Patent No. RE44,326 Issues
2015-09	Comcast announced it would begin to offer its voice recognition system
2016-12-19	Complaint Filing Date

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

U.S. Patent No. 7,260,538 - "Method and Apparatus for Voice Control of a Television Control Device"

The Invention Explained

• Problem Addressed: The patent describes the problem of television remote controls becoming increasingly large and complex due to the addition of more functions and buttons, as well as the high cost of prior art voice recognition systems which required sophisticated processing hardware in close proximity to the user (’538 Patent, col. 1:59-2:6). Another identified problem is the unintentional activation of voice command systems by ambient noise (’538 Patent, col. 2:9-15).
• The Patented Solution: The invention proposes a centralized architecture where a user speaks a voice command into a remote control, which wirelessly transmits the analog or digital voice signal to a set-top box. The set-top box then transmits the signal via the cable network to a remote "head-end unit" where the computationally intensive speech recognition processing occurs. Once the command is recognized, the head-end unit sends a corresponding command function back to the set-top box for execution (’538 Patent, Abstract; Fig. 1). The use of a push-to-talk button on the remote is described as a solution to prevent inadvertent activation by ambient noise (’538 Patent, col. 2:36-39).
• Technical Importance: This system architecture aimed to reduce the cost and complexity of the consumer-premises equipment by centralizing the expensive voice processing function at the network head-end.

Key Claims at a Glance

• The complaint asserts independent claim 1 (Compl. ¶41).
• Essential elements of Claim 1 (method):
  • A television remote control receiving a user-activated indication of a voice command.
  • Receiving the voice command through a microphone associated with the remote.
  • The remote wirelessly transmitting a signal representing the voice command to a cable set-top box.
  • The set-top box transmitting a signal representing the voice command via a cable television link to a remotely located head-end unit.
  • Processing the voice command at the head-end unit.
  • The head-end unit deriving and transmitting a set-top-box-compatible command function back to the set-top box.
  • The set-top box performing the command function.
• The complaint reserves the right to assert dependent claims 2, 3, 4, 6, 18-24, 33-35, 40, and 41 (Compl. ¶43).

U.S. Patent No. 7,047,196 - "System and Method of Voice Recognition Near a Wireline Node of a Network Supporting Cable Television and/or Video Delivery"

The Invention Explained

• Problem Addressed: The patent notes that existing cable television networks provide limited upstream bandwidth from the subscriber to the headend, which restricts the ability to support new interactive services that require significant data transmission from many users, such as centralized voice recognition (’196 Patent, col. 3:5-24, 8:65-9:2).
• The Patented Solution: The invention describes a system and method for utilizing the network's "back channel" to carry voice commands from multiple user sites to a speech processing system located at a "wireline node" (e.g., a headend). The core method involves receiving the back channel, "partitioning" it into a "multiplicity of received identified speech channels," processing those channels to determine the speech content, and generating a unique response for each user based on that content (’196 Patent, Abstract; Fig. 10).
• Technical Importance: This technology describes an architecture for managing and processing voice command data from numerous subscribers over the constrained upstream "back channel" of a hybrid fiber-coaxial cable network.

Key Claims at a Glance

• The complaint asserts independent claim 2 (Compl. ¶49).
• Essential elements of Claim 2 (method):
  • A method for providing voice recognition services.
  • Receiving a back channel to create a received back channel.
  • Partitioning the received back channel into a multiplicity of received identified speech channels, where at least one channel has an associated user site.
  • Processing the speech channels from the associated user site to create a multiplicity of identified speech contents.
  • Responding to the identified speech content from the associated user site to create an identified speech content response for that user site.
• The complaint reserves the right to assert dependent claims 4, 15, 54, and 66 (Compl. ¶51).

U.S. Reissue Patent No. RE44,326 - "System and Method of Voice Recognition Near a Wireline Node of a Network Supporting Cable Television and/or Video Delivery"

Technology Synopsis

This patent is a reissue of a patent from the same family as the ’196 Patent and addresses the same technical challenge: implementing a centralized voice recognition service for many users over a cable network's limited-bandwidth upstream "back channel." The patented solution similarly involves receiving and partitioning this back channel into distinct speech channels at a central wireline node, processing the speech, and generating user-specific responses.

Asserted Claims

Independent claim 11 is asserted (Compl. ¶57).

Accused Features

The complaint accuses Comcast's system of sending subscriber voice commands from a remote to Comcast and its service provider for processing via Automatic Speech Recognition (ASR) and Natural Language Processing (NLP) modules, which then generate an action to be performed (Compl. ¶58).

III. The Accused Instrumentality

Product Identification

The accused instrumentalities are Comcast's XFINITY® and X1 Entertainment Operating Systems® (Compl. ¶40). This includes the system's components, such as voice-enabled remote controls and set-top boxes, as well as the underlying network infrastructure and methods for processing voice commands (Compl. ¶¶ 40, 48, 56).

Functionality and Market Context

The accused systems provide an interactive television platform allowing subscribers to control the service and search for content using voice commands spoken into a remote control (Compl. ¶42.b). According to the complaint, these voice commands are transmitted from the user's equipment to Comcast and its service provider for processing (Compl. ¶58.a). This processing allegedly involves an "ASR module" to convert speech to text and an "NLP [Natural Language Processing] module" to interpret the user's request and determine an action (Compl. ¶58.b). The complaint asserts that Comcast is one of the nation's largest providers of video services and that the accused voice feature is offered through its primary XFINITY® and X1 brands (Compl. ¶12). No probative visual evidence provided in complaint.

IV. Analysis of Infringement Allegations

U.S. Patent No. 7,260,538 Infringement Allegations

Claim Element (from Independent Claim 1)	Alleged Infringing Functionality	Complaint Citation	Patent Citation
a television remote control receiving user-activated indication of a voice command	The accused system includes a remote that allows subscribers to "use simple voice commands."	¶42.b	col. 4:36-37
receiving said voice command through a microphone associated with said television remote control	The accused system includes a remote control into which a subscriber speaks voice commands.	¶58.a	col. 4:25-28
said television remote control wirelessly transmitting a signal representing said voice command to a cable set-top box	The accused remote is paired with a "set-top box" for control of the TV.	¶42.c	col. 4:28-31
said cable set-top box transmitting a signal representing said voice command via cable television link to a remotely located head-end unit	The accused system architecture includes a "head-end" where "content is aggregated and disseminated."	¶42.a	col. 4:7-10
processing said voice command at said head-end unit	Voice commands are sent to Comcast and its service provider for processing.	¶58.a	col. 4:10-12
the head-end unit deriving a . . . command function corresponding to said voice command	An Action Recognition module assembles the processed request into a structure passed for execution.	¶58.b	col. 4:20-24
the head-end unit transmitting said command function to said cable set-top box via the cable television link	After processing at the head-end, a command function is transmitted back to the set-top box.	¶41	col. 2:32-34
performing said command function at said cable set-top box	The set-top box performs the command sent from the head-end.	¶41	col. 2:34-36

U.S. Patent No. 7,047,196 Infringement Allegations

Claim Element (from Independent Claim 2)	Alleged Infringing Functionality	Complaint Citation	Patent Citation
receiving a back channel to create a received back channel	The accused system uses fiber optic and coaxial cable to send and receive signals to/from nodes. Nodes receive upstream RF signals from subscribers via the coaxial infrastructure.	¶50.a-b	col. 6:6-7
partitioning said received back channel into a multiplicity of received identified speech channels, wherein at least one of said identified speech channels has an associated user site	The accused system manages signals to/from nodes located in neighborhoods for transmission back to Comcast's headend or hub.	¶50.a-b	col. 6:8-10
processing said multiplicity of said received identified speech channels from said associated user site to create a multiplicity of said identified speech contents	Voice commands are sent to Comcast for processing to provide the voice control service.	¶58.a	col. 6:11-13
responding to said identified speech content from said associated user site to create an identified speech content response for said associated user site	Comcast provides content consistent with the spoken voice to the customer's television.	¶13	col. 6:14-16

Identified Points of Contention

• Scope Questions: The infringement theory for the ’538 patent may depend on whether Comcast's current architecture, which the complaint alleges involves "Comcast and its contracted service provider for processing" (Compl. ¶58.a), meets the claim limitation of a "cable television head-end unit." A potential dispute is whether this term, as used in the patent, is limited to a traditional, singular cable company facility or can be construed to cover a distributed, cloud-based, or third-party-hosted processing system.
• Technical Questions: For the ’196 and ’326 patents, a central technical question may be whether Comcast's method for handling upstream voice traffic from multiple users constitutes "partitioning" a "back channel" into "speech channels" as claimed. The complaint describes the physical network infrastructure (Compl. ¶50.a-b) but provides less detail on the specific protocol used to manage concurrent voice data streams. The analysis may turn on whether Comcast's modern packet-switching or multiplexing technology performs the function of the claimed "partitioning" or operates in a fundamentally different way.

V. Key Claim Terms for Construction

• The Term: "cable television head-end unit" (’538 Patent, Claim 1)
• Context and Importance: This term is critical because the claim requires specific processing steps to occur "at said head-end unit." The complaint alleges processing by "Comcast and its contracted service provider" (Compl. ¶58.a). The viability of the infringement claim may depend on whether this distributed, potentially third-party architecture falls within the scope of a "head-end unit."
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The specification refers to the head-end unit functionally as a "central processing station" (’538 Patent, col. 4:16-17) that is "designed to handle a multitude of voice command inputs from a multitude of cable television users" (’538 Patent, col. 4:17-19). This language may support an interpretation based on the function performed (centralized processing for multiple cable users) rather than a specific physical location or ownership structure.
  • Evidence for a Narrower Interpretation: The patent's figures and description consistently place the "head-end unit" within a traditional cable television network diagram, distinct from the subscriber's "set-top box" and "remote control" (’538 Patent, Fig. 1). This context could support a narrower construction limited to the conventional meaning of a cable system's primary signal processing and distribution facility.
• The Term: "partitioning said received back channel into a multiplicity of received identified speech channels" (’196 Patent, Claim 2)
• Context and Importance: This term defines the core technical step for managing voice data from multiple users. Infringement will depend on whether Comcast's network protocol for handling upstream traffic performs this "partitioning" function. Practitioners may focus on this term because modern IP-based networks may handle data streams in ways that differ from the channelization concepts prevalent when the patent was filed.
• Intrinsic Evidence for Interpretation:
  • Evidence for a Broader Interpretation: The patent describes this step functionally in its summary and abstract, without limiting it to a specific technology like time-division or frequency-division multiplexing (’196 Patent, Abstract). The term "partitioning" could be argued to broadly cover any logical separation of user data streams from a shared upstream medium.
  • Evidence for a Narrower Interpretation: The background section discusses the limitations of the Aloha protocol and fixed upstream channels in early digital cable systems (’196 Patent, col. 3:5-40). This context may support an argument that "partitioning" into "channels" refers to a specific type of resource allocation on the physical communication layer, as opposed to the routing of discrete data packets in a modern IP network.

VI. Other Allegations

• Indirect Infringement: The complaint does not plead separate counts for indirect infringement. The claims for relief focus on direct infringement under 35 U.S.C. § 271(a). However, the factual allegations state that Comcast provides subscribers with devices like remotes and set-top boxes and that its publications instruct users to "use simple voice commands," which may form a basis for later claims of induced infringement (Compl. ¶42.b).
• Willful Infringement: Willfulness is alleged for all three patents-in-suit (Compl. ¶¶ 44, 52, 59). The allegations are based on extensive pre-suit knowledge stemming from a "confidential relationship" beginning around 2001, during which Promptu allegedly disclosed its technology, patents, and patent applications to Comcast long before Comcast launched its own voice recognition system in 2015 (Compl. ¶¶ 15-21, 37).

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

• A core issue will be one of architectural scope: does Comcast’s modern, potentially cloud-based and third-party-supported voice processing system constitute a "cable television head-end unit" and "wireline node" as those terms are used in the context of the asserted patents, or does its distributed nature place it outside the claimed architecture?
• A key technical question will be one of protocol equivalence: does the packet-based method by which Comcast’s network manages simultaneous upstream voice commands from numerous users perform the claimed function of "partitioning" a "back channel" into distinct "speech channels," or is there a fundamental mismatch in the underlying network technology?
• A central question for damages will be willfulness: based on the extensive alleged history of disclosures and collaboration between the parties, did Comcast's development and launch of its voice control platform constitute objective recklessness with regard to Plaintiff's patent rights, potentially justifying an award of enhanced damages?