Systems and methods for determining and using interaction models
First Claim
1. A method of determining user interactions comprising:
- determining speech information;
determining discourse functions and prosodic features in the speech information;
selecting a predictive interaction model, the predictive interaction model developed by associating discourse functions and prosodic features of a training corpus of turn annotated speech with turn information of the training corpus of turn annotated speech; and
determining an interaction turn, in the speech information, based on the predictive interaction model and the discourse functions of the speech information and the prosodic features of the speech informationwherein the predictive interaction model accepts the prosodic features of the speech information as one input and a current discourse function of the speech information as another input and determines a likelihood that a next discourse function is associated with the interaction turn.
1 Assignment
0 Petitions
Accused Products
Abstract
Techniques are provided for determining and using interaction models. Discourse functions, prosodic features and turn information are determined from the speech information in a training corpus. Statistics, decision trees, rules and/or various other methods are used to determine a predictive interaction model based on the discourse functions, the prosodic features and the turn information. Predictive interaction models are optionally determined for individual users, genres, languages and/or other characteristics of the speech information. The predictive interaction model is useable to predict turns in a dialogue based on the discourse functions and prosodic features identified in the speech information. Speech information is presented and/or received based on the predictive interaction model.
-
Citations
20 Claims
-
1. A method of determining user interactions comprising:
-
determining speech information; determining discourse functions and prosodic features in the speech information; selecting a predictive interaction model, the predictive interaction model developed by associating discourse functions and prosodic features of a training corpus of turn annotated speech with turn information of the training corpus of turn annotated speech; and determining an interaction turn, in the speech information, based on the predictive interaction model and the discourse functions of the speech information and the prosodic features of the speech information wherein the predictive interaction model accepts the prosodic features of the speech information as one input and a current discourse function of the speech information as another input and determines a likelihood that a next discourse function is associated with the interaction turn. - View Dependent Claims (2, 3, 4, 12, 13, 15, 16, 17)
-
-
5. A method of determining a predictive interaction model comprising:
-
determining a training corpus of turn annotated speech information; determining discourse functions and prosodic features associated with a turn information of the turn annotated speech information; and determining a predictive interaction model based on the discourse functions, the prosodic features and the turn information, wherein the predictive interaction model is adapted to accepting the prosodic features of a speech information as one input and a current discourse function of the speech information as another input and determining a likelihood that a next discourse function is associated with the turn information. - View Dependent Claims (6, 14, 18)
-
-
7. A system for determining interactions comprising:
-
an input/output circuit for retrieving recognized speech and prosodic features; a processor that determines speech information; and
discourse functions and prosodic features in the speech information;
determines a predictive interaction model; and
determines an interaction turn based on the predictive interaction model and the discourse functions and prosodic features in the speech informationwherein the predictive interaction model accepts the prosodic features of the speech information as one input and a current discourse function of the speech information as another input and determines a likelihood that a next discourse function of the speech information is associated with the interaction turn. - View Dependent Claims (8, 9, 10, 19, 20)
-
-
11. A computer readable storage medium embodying a set of instructions, the set of instructions, when executed by one or more processors, causing the one or more processors perform a method for determining interactions, the method comprising:
-
determining speech information; determining discourse functions and prosodic features in the speech information; determining a predictive interaction model, the predictive interaction model being developed by associating discourse functions of a training corpus of turn annotated speech and prosodic features of the turn annotated speech with turn information of the turn annotated speech; and determining an interaction turn based on the predictive interaction model and the determined discourse functions and prosodic features, wherein the predictive interaction model accepts the prosodic features of the speech information as one input and a current discourse functions of the speech information as another input and determines a likelihood that a next discourse function is associated with the interaction turn.
-
Specification