US 8073681

The search for US patent 8073681 on Google Patents (which typically reflects USPTO data) provides the following information:

US Patent 8073681 Summary:

• Title: System and method for a cooperative conversational voice user interface
• Assignee: VB Assets LLC (Current Assignee) (Originally assigned to VoiceBox Technologies Corp)
• Inventors: Larry Baldwin, Tom Freeman, Michael Tjalve, Blane Ebersold, Chris Weider
• Filing Date: 2006-10-16
• Issue Date: 2011-12-06
• Abstract: A cooperative conversational voice user interface is provided. The cooperative conversational voice user interface may build upon short-term and long-term shared knowledge to generate one or more explicit and/or implicit hypotheses about an intent of a user utterance. The hypotheses may be ranked based on varying degrees of certainty, and an adaptive response may be generated for the user. Responses may be worded based on the degrees of certainty and to frame an appropriate domain for a subsequent utterance. In one implementation, misrecognitions may be tolerated, and conversational course may be corrected based on subsequent utterances and/or responses.

Plain-Language Overview of Independent Claims:

The patent text contains 42 claims, I will identify the independent ones (typically claims 1, 10, and 20 based on structure, or claims that don't refer to other claims) and then summarize them.

• Claim 1 (System Claim): This claim describes a system for a cooperative conversational voice user interface. It includes an input device to receive a human utterance, a speech recognition engine to generate preliminary interpretations, and a conversational speech engine. The conversational speech engine comprises a free-form voice search module (to understand natural language and account for speaking variations), a noise tolerance module (to filter irrelevant words/noise), and a context determination process (to establish conversation meaning using competing context domain agents). This engine uses short-term and long-term shared knowledge to generate hypotheses about user intent and create adaptive responses that can evolve the conversation.
• Claim 10 (Method Claim): This claim describes a method for a cooperative conversational voice user interface. It involves receiving a human utterance with one or more requests, generating preliminary interpretations, and processing these interpretations with a conversational speech engine. The processing includes understanding free-form language (accounting for variations, jargon, word order, pauses, and imperfect speech), filtering noise, determining context using competing domain agents, building hypotheses about user intent using shared knowledge, and generating adaptive responses. These responses consider factors like contextual signifiers, grammatical rules, response statistics, and aim to influence subsequent user replies for easier recognition, while also correcting conversational flow after misrecognitions.
• Claim 19 (System Claim): This claim describes a system that includes a speech recognition engine and a conversational speech engine to generate an adaptive conversational response. The conversational speech engine includes a free-form voice search module that understands natural language, tolerates variations, and infers requests from incomplete or contradictory information. It also has a noise tolerance module to filter out noise and irrelevant words and a context determination process that uses competing context domain agents to establish meaning and update shared knowledge.

CAFC 2026 Dockets Search for US8073681:

A review of the search results for CAFC 2026 dockets did not specifically mention US patent 8073681. The results provided general updates on CAFC patent cases in April and May 2026, including discussions on patent eligibility, written description, and infringement cases involving various companies (e.g., Google, Eli Lilly, Columbia University) and other patent numbers (e.g., US11244675B2, US9223487B2, US11069337B2, US8296383B2). There is a Google Patents entry that shows "Family has litigation" and lists several court cases, including some in the Court of Appeals for the Federal Circuit (e.g., case 25-1113, 25-1142, 25-1357, 25-1854). However, the provided search results from specific CAFC dockets for 2026 did not directly yield specific information within the search snippets linking these cases directly to US8073681 for the year 2026. Therefore, based on the provided search results, direct confirmation of ongoing CAFC litigation specifically mentioning the patent number 8073681 in the provided 2026 docket snippets is uncertain. More in-depth investigation into each of those listed CAFC cases would be required to confirm the status for 2026.US patent 8073681, titled "System and method for a cooperative conversational voice user interface," was filed on October 16, 2006, and issued on December 6, 2011. The current assignee is VB Assets LLC, though it was originally assigned to VoiceBox Technologies Corp. The inventors are Larry Baldwin, Tom Freeman, Michael Tjalve, Blane Ebersold, and Chris Weider.

Abstract:
The patent describes a cooperative conversational voice user interface that leverages short-term and long-term shared knowledge to formulate explicit and/or implicit hypotheses about a user's intent from an utterance. These hypotheses are ranked by certainty, and an adaptive response is generated. The responses are designed based on these certainty levels and aim to guide subsequent user utterances. The system is designed to tolerate misrecognitions and correct the conversational flow through ongoing interaction.

Plain-Language Overview of Independent Claims:

• Claim 1 (System for a cooperative conversational voice user interface): This claim describes a system that receives a human utterance through an input device. A speech recognition engine processes this utterance into preliminary interpretations. A conversational speech engine further processes these interpretations, featuring a free-form voice search module (to understand natural language, jargon, varying word order, and imperfect speech), a noise tolerance module (to filter out irrelevant words and environmental noise), and a context determination process (which uses competing context domain agents to establish conversation meaning and update shared knowledge). The system uses both short-term (within a conversation) and long-term (user-centric, over time) shared knowledge to form hypotheses about user intent and generate adaptive responses that can steer the conversation forward.
• Claim 10 (Method for a cooperative conversational voice user interface): This claim outlines a method involving receiving a human utterance and generating preliminary interpretations. These interpretations are then processed by a conversational speech engine. The method includes understanding free-form language (including jargon, slang, varied word order, verbalized pauses, and imperfect speech), filtering out noise (both verbal and environmental), and determining the context of the request through competing context domain agents. It further involves building hypotheses about the user's intent using accumulated shared knowledge (both short-term and long-term) and generating adaptive responses. These responses are designed to be sensitive to context and grammar, statistically rated, and even randomized to create a natural conversational feel. They also frame questions to influence user replies for easier recognition and correct conversational missteps through clarification rather than explicit error messages.
• Claim 19 (System for generating an adaptive conversational response): This claim focuses on a system that generates an adaptive conversational response. It comprises a speech recognition engine and a conversational speech engine. The conversational speech engine includes a free-form voice search module (capable of understanding natural language, tolerating variations, and inferring requests from incomplete or contradictory information), a noise tolerance module (for discarding meaningless words or noise), and a context determination process (which determines conversational context using competing context domain agents and updates shared short-term and long-term shared knowledge).

CAFC 2026 Dockets:
While Google Patents indicates that US8073681 has "Family has litigation" and lists several cases in the Court of Appeals for the Federal Circuit, the provided search snippets for specific 2026 CAFC dockets do not directly confirm or detail ongoing litigation specifically mentioning patent number 8073681 in 2026. Further investigation into each of the listed CAFC cases would be necessary to determine their current status and relevance to US8073681 for the year 2026.