Methods and apparatus for generating dialog state conditioned language models

• US 7,853,449 B2
• Filed: 03/28/2008
• Issued: 12/14/2010
• Est. Priority Date: 03/27/2002
• Status: Expired due to Term

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Visweswariah, Karthik, Monkowski, Michael Daniel, Dharanipragada, Satyanarayana, Printz, Harry W.
• Primary Examiner(s): Lerner; Martin
• Application Number:
  US12/057,646
• Publication Number:
  US 20080215329A1
• Time in Patent Office: 991 Days
• Field of Search: 704/238, 704/245, 704/255, 704/256.3, 704/275, 704/257
• US Class Current: 704/238
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models
• CPC Class Codes: G10L 15/197 Probabilistic grammars, e.g. word n-grams
• CPC Class Codes: G10L 2015/228 of application context

Abstract

Techniques are provided for generating improved language modeling. Such improved modeling is achieved by conditioning a language model on a state of a dialog for which the language model is employed. For example, the techniques of the invention may improve modeling of language for use in a speech recognizer of an automatic natural language based dialog system. Improved usability of the dialog system arises from better recognition of a user'"'"'s utterances by a speech recognizer, associated with the dialog system, using the dialog state-conditioned language models. By way of example, the state of the dialog may be quantified as: (i) the internal state of the natural language understanding part of the dialog system; or (ii) words in the prompt that the dialog system played to the user.

Claims

1. A method for use in accordance with a dialog system, the method comprising:

generating at least one language model, the at least one language model being conditioned on a state of dialog associated with the dialog system; and

storing the at least one language model for subsequent use in accordance with a speech recognizer associated with the dialog system;

wherein generating the at least one language model conditioned on a state of dialog associated with the dialog system further comprises;

dividing training data which is labeled by state into different state sets depending on the state to which the training data belongs;

clustering the state sets into clustered state sets, the clustering comprising combining training data belonging to states that are close to each other based on a distance measure;

building a separate language model for each of the clustered state sets to create a plurality of separate language models; and

building at least one interpolated model by interpolating one or more of the plurality of separate language models with a base model obtained from training data that includes at least some training data used to train all of the plurality of separate language models.

2. The method of claim 1, wherein the dialog system is a natural language understanding based dialog system.

3. The method of claim 2, wherein the state corresponds to an internal state of the natural language understanding part of the dialog system.

4. The method of claim 1, wherein the state corresponds to a prompt that the dialog system presents to a user.

5. The method of claim 1, wherein each separate language model is a trigram language model.

6. The method of claim 1, wherein each separate language model is built using a modified Kneser-Ney smoothing technique.

7. The method of claim 1, wherein each separate language model is interpolated with a base model obtained from available training data for a domain of the dialog system.

8. The method of claim 1, wherein a decision to cluster two states is based on a distance measure computed between respective word distributions associated with the two states.

9. Apparatus for use in accordance with a dialog system, the apparatus comprising:

at least one processor operative to generate at least one language model, the at least one language model being conditioned on a state of dialog associated with the dialog system; and

memory, coupled to the at least one processor, for storing the at least one language model for subsequent use in accordance with a speech recognizer associated with the dialog system;

wherein the operation of generating the at least one language model conditioned on a state of dialog associated with the dialog system further comprises;

(i) dividing training data which is labeled by state into different state sets depending on the state to which the training data belongs;

(ii) clustering the state sets into clustered state sets, the clustering comprising combining training data belonging to states that are close to each other based on a distance measure;

(iii) building a separate language model for each of the clustered state sets to create a plurality of separate language models; and

(iv) building at least one interpolated model by interpolating one or more of the plurality of separate language models with a base model obtained from training data that includes at least some training data used to train all of the plurality of separate language models.

10. The apparatus of claim 9, wherein the dialog system is a natural language understanding based dialog system.

11. The apparatus of claim 10, wherein the state corresponds to an internal state of the natural language understanding part of the dialog system.

12. The apparatus of claim 9, wherein the state corresponds to a prompt that the dialog system presents to a user.

13. At least one memory device storing instructions that, when executed by at least one processor, perform a method for use in accordance with a dialog system, the method comprising:

generating at least one language model, the at least one language model being conditioned on a state of dialog associated with the dialog system; and

storing the at least one language model for subsequent use in accordance with a speech recognizer associated with the dialog system;

wherein generating the at least one language model conditioned on a state of dialog associated with the dialog system further comprises;

dividing training data which is labeled by state into different state sets depending on the state to which the training data belongs;

clustering the state sets into clustered state sets, the clustering comprising combining training data belonging to states that are close to each other based on a distance measure;

building a separate language model for each of the clustered state sets to create a plurality of separate language models; and

building at least one interpolated model by interpolating one or more of the plurality of separate language models with a base model obtained from training data that includes at least some training data used to train all of the plurality of separate language models.