System and method for dialog modeling

Patent metadata

Field	Value
Publication number	US-9972307-B2
Application number	US-201514845634-A
Country	US
Kind code	B2
Filing date	Sep 4, 2015
Priority date	Nov 26, 2008
Publication date	May 15, 2018
Grant date	May 15, 2018

Abstract

Disclosed herein are systems, computer-implemented methods, and computer-readable media for dialog modeling. The method includes receiving spoken dialogs annotated to indicate dialog acts and task/subtask information, parsing the spoken dialogs with a hierarchical, parse-based dialog model which operates incrementally from left to right and which only analyzes a preceding dialog context to generate parsed spoken dialogs, and constructing a functional task structure of the parsed spoken dialogs. The method can further either interpret user utterances with the functional task structure of the parsed spoken dialogs or plan system responses to user utterances with the functional task structure of the parsed spoken dialogs. The parse-based dialog model can be a shift-reduce model, a start-complete model, or a connection path model.

First claim

We claim:

1. A method comprising:
training a plurality of hierarchical, parsed-based dialog models, wherein each of the plurality of hierarchical, parsed-based dialog models operates incrementally from left to right and only analyzes an immediately preceding dialog context and wherein the plurality of hierarchical, parsed-based dialog models comprises one of a shift-reduce model, a start-complete model or a connection path model, and wherein:
when the plurality of hierarchical, parsed-based dialog models comprises a shift-reduce model, the shift-reduce model has a stack and a tree which (a) shifts each utterance onto the stack, (b) inspects the stack, and (c) based on a stack inspection, performs a reduce action that creates subtrees in the tree;
when the plurality of hierarchical, parsed-based dialog models comprises a start-complete model, the start-complete model uses a stack to maintain a global parse state and produces a dialog task structure directly without producing an equivalent tree; and
when the plurality of hierarchical, parsed-based dialog models comprises a connection path model, the connection path model does not use a stack to maintain a global parse state, and wherein the connection path model (a) directly predicts a connection path from a root to a terminal for each received spoken dialog, and (b) creates a parse tree representing the connection path for each received spoken dialog;

parsing, via a processor, spoken dialogs with a hierarchical, parse-based dialog model from the plurality of hierarchical, parsed-based dialog models, to yield parsed spoken dialogs;
constructing a functional task structure of the parsed spoken dialogs;
predicting a likely next dialog act in a spoken dialog using the functional task structure and the hierarchical, parsed-based dialog model, the likely next dialog act corresponding to a next utterance comprising a clause to be spoken by a speaker, wherein the predicting occurs prior to receiving the next utterance; and
selecting a language model for the next utterance based on the likely next dialog act.

2. The method of claim 1 , further comprising measuring a dialog efficiency at different dialog stages based on the language model selected.

3. A system comprising:
a processor; and
a computer-readable storage medium having instructions stored which, when executed by the processor, cause the processor to perform operations comprising:
training a plurality of hierarchical, parsed-based dialog models wherein the plurality of hierarchical, parsed-based dialog models operate incrementally from left to right and only analyze an immediately preceding dialog context and wherein the plurality of hierarchical, parsed-based dialog models comprises one of a shift-reduce model, a start-complete model or a connection path model, and wherein:
when the plurality of hierarchical, parsed-based dialog models comprises a shift-reduce model, the shift-reduce model has a stack and a tree which (a) shifts each utterance onto the stack, (b) inspects the stack, and (c) based on a stack inspection, performs a reduce action that creates subtrees in the tree;
when the plurality of hierarchical, parsed-based dialog models comprises a start-complete model, the start-complete model uses a stack to maintain a global parse state and produces a dialog task structure directly without producing an equivalent tree; and
when the plurality of hierarchical, parsed-based dialog models comprises a connection path model, the connection path model does not use a stack to maintain a global parse state, and wherein the connection path model (a) directly predicts a connection path from a root to a terminal for each received spoken dialog, and (b) creates a parse tree representing the connection path for each received spoken dialog;
parsing spoken dialogs with a hierarchical, parse-based dialog model from the plurality of hierarchical, parsed-based dialog models, to yield parsed spoken dialogs;
constructing a functional task structure of the parsed spoken dialogs;
predicting a likely next dialog act in a spoken dialog using the functional task structure and the hierarchical, parsed-based dialog model, the likely next dialog act corresponding to a next utterance comprising a clause to be spoken by a speaker, wherein the predicting occurs prior to receiving the next utterance; and
selecting a language model for the next utterance based on the likely next dialog act.

4. The system of claim 3 , the computer-readable storage medium having additional instructions stored which, when executed by the processor, cause the processor to perform operations comprising measuring a dialog efficiency at different dialog stages based on the language model selected.

5. A computer-readable storage device having instructions stored which, when executed by a computing device, cause the computing device to perform operations comprising:
training a plurality of hierarchical, parsed-based dialog models wherein the plurality of hierarchical, parsed-based dialog models operate incrementally from left to right and only analyze an immediately preceding dialog context and wherein the plurality of hierarchical, parsed-based dialog models comprises one of a shift-reduce model, a start-complete model or a connection path model, and wherein:
when the plurality of hierarchical, parsed-based dialog models comprises a shift-reduce model, the shift-reduce model has a stack and a tree which (a) shifts each utterance onto the stack, (b) inspects the stack, and (c) based on a stack inspection, performs a reduce action that creates subtrees in the tree;
when the plurality of hierarchical, parsed-based dialog models comprises a start-complete model, the start-complete model uses a stack to maintain a global parse state and produces a dialog task structure directly without producing an equivalent tree; and
when the plurality of hierarchical, parsed-based dialog models comprises a connection path model, the connection path model does not use a stack to maintain a global parse state, and wherein the connection path model (a) directly predicts a connection path from a root to a terminal for each received spoken dialog, and (b) creates a parse tree representing the connection path for each received spoken dialog;

parsing spoken dialogs with a hierarchical, parse-based dialog model from the plurality of hierarchical, parsed-based dialog models, to yield parsed spoken dialogs;
constructing a functional task structure of the parsed spoken dialogs;
predicting a likely next dialog act in a spoken dialog using the functional task structure and the hierarchical, parsed-based dialog model, the likely next dialog act corresponding to a next utterance comprising a clause to be spoken by a speaker, wherein the predicting occurs prior to receiving the next utterance; and
selecting a language model for the next utterance based on the likely next dialog act.

Assignees

• At & T Ip I Lp

Inventors

• Stent Amanda
• Bangalore Srinivas

Classifications

• G06F40/137

  Hierarchical processing, e.g. outlines · CPC title

• G10L15/04

  Segmentation; Word boundary detection · CPC title

• G10L2015/0638

  Interactive procedures · CPC title

• G10L25/12

  the extracted parameters being prediction coefficients · CPC title

• G06F40/12

  Use of codes for handling textual entities · CPC title