System and method for a cooperative conversational voice user interface

• US 8,515,765 B2
• Filed: 10/03/2011
• Issued: 08/20/2013
• Est. Priority Date: 10/16/2006
• Status: Active Grant

• Current Assignee: VB Assets, LLC
• Original Assignee: VoiceBox Technologies, Inc. (Microsoft Corporation)
• Inventors: Baldwin, Larry, Freeman, Tom, Tjalve, Michael, Ebersold, Blane, Weider, Chris
• Primary Examiner(s): YEN, ERIC L
• Application Number:
  US13/251,712
• Publication Number:
  US 20120022857A1
• Time in Patent Office: 687 Days
• Field of Search: 704270-275, 704/9
• US Class Current: 704/275
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G06F 40/30 Semantic analysis
• CPC Class Codes: G10L 15/18 using natural language modelling
• CPC Class Codes: G10L 15/1815 Semantic context, e.g. disambiguation of the recognition hypotheses based on word meaning
• CPC Class Codes: G10L 15/1822 Parsing for meaning understanding
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: G10L 2015/0631 Creating reference templates; Clustering
• CPC Class Codes: G10L 2015/225 Feedback of the input speech
• CPC Class Codes: G10L 2015/228 of application context
• CPC Class Codes: G10L 2021/02166 Microphone arrays; Beamforming
• CPC Class Codes: G10L 25/51 for comparison or discrimination
• CPC Class Codes: G10L 25/63 for estimating an emotional state

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.

Claims

1. A method of providing a voice interface, comprising:

receiving, at a speech engine, a natural language utterance from a voice-enabled device, the natural language utterance corresponding to a conversation type;

determining the conversation type corresponding to the natural language utterance based on whether a user that spoke the natural language utterance has a leader role in an interaction with the voice-enabled device or has a supporter role in the interaction with the voice-enabled device; and

generating a response to the natural language utterance with a format based on the conversation type, wherein the format is adapted to limit the user'"'"'s future input to interjecting queries or requests for clarification if the user has the supporter role.

2. The method recited in claim 1, wherein determining the conversation type comprises determining a conversational goal corresponding to the natural language utterance, wherein the format of the response is adapted to provide a discrete piece of information to the user, to provide a plurality of pieces of related information to the user, or to obtain one or more pieces of information from the user.

3. The method recited in claim 2, wherein the response generated in the format that is adapted to provide the discrete piece of information is shorter than the response generated in the format that is adapted to provide the plurality of pieces of related information.

4. The method recited in claim 1, wherein determining the conversation type includes determining a conversational goal corresponding to the natural language utterance, wherein the conversation type is determined as a query conversation if the conversational goal is to gather discrete information or perform a discrete task, is determined as a didactic conversation if the conversational goal is to gather related information, or is determined as an exploratory conversation if the natural language utterance has no specific goal.

5. The method recited in claim 1, wherein the conversation type is determined as a query conversation if the user that spoke the natural language utterance has the leader role, is determined as a didactic conversation if the user that spoke the natural language utterance has the supporter role, or is determined as an exploratory conversation if the user that spoke the natural language utterance has both the leader role and the supporter role.

6. The method recited in claim 1, wherein determining the conversation type is based on an allocation of information between the user and the speech engine, wherein the conversation type is determined to be a query conversation or a didactic conversation if the user that spoke the natural language utterance has a lower allocation of information related to the natural language utterance than the speech engine, or is determined as a didactic conversation if the user and the speech engine have an equal or alternating allocation of information related to the natural language utterance.

7. The method recited in claim 1, wherein determining the conversation type further comprises determining an allocation of information between the user and the speech engine.

8. The method recited in claim 1, wherein generating the response comprises generating, if the conversation type is a query conversation, a response that only provides information requested by the natural language utterance, and generating, if the conversation type is a didactic conversation, a response that controls the conversation and limits the user'"'"'s future input to a request or query for clarification.

9. The method recited in claim 1, wherein the natural language utterance is part of a conversation that comprises related natural language utterances from the user, wherein determining whether the user has the leader role or the supporter role comprises determining whether the user controls the conversation or whether the user provides information to the speech engine during the conversation.

10. A system for providing a voice interface, comprising:

a speech engine configured to receive a natural language utterance from a voice-enabled device, the natural language utterance corresponding to a conversation type, wherein the speech engine includes a processor configured to;

determine the conversation type corresponding to the natural language utterance based on whether a user that spoke the natural language utterance has a leader role in an interaction with the voice-enabled device or has a supporter role in the interaction with the voice-enabled device;

a response builder configured to generate a response to the natural language utterance with a format based on the conversation type, wherein the format is adapted to limit the user'"'"'s future input to interjecting queries or requests for clarification if the user has the supporter role.

11. The system recited in claim 10, wherein the speech engine is configured to determine the conversation type further based on a conversational goal corresponding to the natural language utterance, and wherein the format of the response is adapted to provide a discrete piece of information to the user, to provide a plurality of pieces of related information to the user, or to obtain one or more pieces of information from the user.

12. The system recited in claim 10, wherein the speech engine is configured to determine the conversation type further based on a conversational goal corresponding to the natural language utterance, and wherein the conversation type is determined to be a query conversation if the conversational goal is to gather discrete information or perform a discrete task, is determined to be a didactic conversation if the conversational goal is to gather related information, or is determined to be an exploratory conversation if the natural language utterance has no specific goal.

13. The system recited in claim 10, wherein the speech engine is configured to determine the conversation type based on the role of the user, wherein the conversation type is determined to be a query conversation if the user that spoke the natural language utterance has the leader role, is determined to be a didactic conversation if the user that spoke the natural language utterance has the supporter role, or is determined to be an exploratory conversation if the user that spoke the natural language utterance has both the leader role and the supporter role.

14. The system recited in claim 10, wherein the speech engine is configured to determine the conversation type based on an allocation of information between the user and the speech engine, wherein the conversation type is determined to be a query conversation or a didactic conversation if the user that spoke the natural language utterance has a lower allocation of information related to the natural language utterance than the speech engine or is determined to be a didactic conversation if the user and the speech engine have an equal or alternating allocation of information.