System and method for enriching spoken language translation with dialog acts

1. A method comprising:

tagging dialog acts associated with a user utterance in a source natural spoken language using a classification model to yield dialog act tags, the dialog act tags being domain independent descriptions of an intended action a speaker carries out by uttering the user utterance;

producing, via a processor, an enriched hypothesis of the user utterance incorporating the dialog act tags; and

outputting a version of the enriched hypothesis translated into a target natural spoken language to yield a translated speech output signal with a word order based on the dialog act tags.

2. The method of claim 1 , wherein the dialog acts are tagged using tags that are grouped into sets.

3. The method of claim 2 , wherein each of the sets is associated with a dialog act category selected from a group consisting of a statement category, an acknowledgement category, an abandoned category, an agreement category, a question category, and an appreciation category.

4. The method of claim 1 , wherein outputting the version of the enriched hypothesis uses a dialog act specific translation model containing a phrase translation table.

5. The method of claim 4 , the method further comprising:

appending to each phrase translation table associated with a particular dialog act specific translation model those entries from a complete model that are not present in the phrase table of the dialog act specific translation model, to yield appended entries; and

weighting the appended entries.

6. The method of claim 4 , wherein the dialog act specific translation model is a bag-of-words translation model.

7. The method of claim 1 , wherein the user utterance is part of a dialog turn having multiple sentences, the method further comprising:

segmenting the user utterance, to yield segments;

tagging second dialog acts in each segment using a maximum entropy model, to yield second tagged dialog acts; and

producing an enriched hypothesis of each segment incorporating the second tagged dialog acts.

8. The method of claim 1 , the method further comprising annotating tagged dialog acts.

9. The method of claim 1 , wherein the classification model is a maximum entropy model.

10. A system comprising:

a processor; and

a memory storing instructions for controlling the processor to perform a method comprising:

tagging dialog acts associated with a user utterance in a source natural spoken language using a classification model to yield dialog act tags, the dialog act tags being domain independent descriptions of an intended action a speaker carries out by uttering the user utterance;

producing an enriched hypothesis of the user utterance incorporating the dialog act tags; and

outputting a version of the enriched hypothesis translated into a target natural spoken language to yield a translated speech output signal with a word order based on the dialog act tags.

11. The system of claim 10 , wherein the dialog acts are tagged using tags that are grouped into sets.

12. The system of claim 11 , wherein each of the sets is associated with a dialog act category selected from a group consisting of a statement category, an acknowledgement category, an abandoned category, an agreement category, a question category, and an appreciation category.

13. The system of claim 10 , wherein outputting the version of the enriched hypothesis is based on a dialog act specific translation model containing a phrase translation table.

14. The system of claim 13 , the instructions further comprising:

appending to each phrase translation table associated with a particular dialog act specific translation model those entries from a complete model that are not present in the phrase table of the dialog act specific translation model, to yield appended entries; and

weighting the appended entries.

15. The system of claim 13 , wherein the dialog act specific translation model is a bag-of-words translation model.

16. The system of claim 10 , wherein the user utterance is part of a dialog turn having multiple sentences, the instructions further comprising:

segmenting the user utterance, to yield segments;

tagging second dialog acts in each segment using a maximum entropy model, to yield second tagged dialog acts; and

producing an enriched hypothesis of each segment incorporating the second tagged dialog acts.

17. A non-transitory computer-readable medium storing a computer program having instructions for controlling a computing device to perform a method comprising:

tagging dialog acts associated with a user utterance in a source natural spoken language using a classification model to yield dialog act tags, the dialog act tags being domain independent descriptions of an intended action a speaker carries out by uttering the user utterance;

producing, via a processor, an enriched hypothesis of the user utterance incorporating the dialog act tags; and

outputting a version of the enriched hypothesis translated into a target natural spoken language to yield a translated speech output signal with a word order based on the dialog act tags.

18. The non-transitory computer-readable medium of claim 17 , wherein the dialog acts are tagged using tags that are grouped into sets.

19. The non-transitory computer-readable medium of claim 18 , wherein each of the sets is associated with a dialog act category selected from a group consisting of a statement category, an acknowledgement category, an abandoned category, an agreement category, a question category, and an appreciation category.

20. The non-transitory computer-readable medium of claim 17 , wherein outputting the version of the enriched hypothesis uses a dialog act specific translation model containing a phrase translation table.