Automatic generation of voice content for a voice response system

1. A method for building a voice response system, the method comprising:

defining an interaction between the voice response system and a user, the interaction comprising text elements to be played to the user and utterances to be spoken by the user, the text elements and the utterances being arranged in a form of a pseudo-dialog wherein the text elements to be played in response to an utterance are identified, said defining the interaction comprising automatically generating at least some of the text elements by automatically generating topic elements from text of a natural language text document by partitioning the text document into topics based on a frequency of occurrence of nouns therein, wherein each topic element comprises a subsection of the text document that relates to a particular topic; and

automatically generating code that uses the interaction to create a voice user-interface.

2. The method of claim 1 , further comprising generating keywords for each topic based on a frequency of occurrence of words in the text of a topic, the code then comprising instructions to select that topic to be played which has a keyword that matches an utterance of the user.

3. The method of claim 2 , wherein the utterance of the user comprises a word pattern that includes the keyword.

4. The method of claim 2 , wherein the text elements comprise navigation elements that include disambiguating questions to be played to the user in the event of the word pattern uttered by the user matching a keyword for more than one topic.

5. The method of claim 4 , wherein automatically generating at least some of the text elements comprises automatically generating the navigation elements, wherein each disambiguating question is based on the most frequently occurring mutually exclusive noun from the text of the topics that have matching keywords.

6. The method of claim 1 , wherein the text elements comprised prompts to be played to the user, the method further comprising rewriting the prompt to increase a likelihood of it being understood by the user.

7. The method of claim 6 , wherein the rewriting comprises rewriting the prompt so that it is in the active tense.

8. The method of claim 6 , wherein rewriting each prompt comprises replacing ambiguous and less frequently used words therein with clear and more frequently used words.

9. The method of claim 6 , wherein the rewriting comprises shortening the prompt if it exceeds a predetermined number of words.

10. The method of claim 6 , further comprising analyzing each prompt to determine a probability of an utterance by the user in response to the prompt being recognized; and notifying an operator if the probability is low.

11. The method of claim 10 , wherein the analyzing comprises determining a phonetic composition of choices presented to the user in each prompt in order to determine a likelihood of the choices being mis-recognized.

12. A method for building a voice response system, the method comprising:

developing voice content for the voice response system, the voice content including prompts and information to be played to a user, said developing comprising reading a natural language text document and automatically generating topics therefrom corresponding to the information to be played to the user based on a frequency of occurrence of nouns in the text document; and

integrating the voice content with logic to define a voice user-interface that is capable of interacting with the user in a manner of a conversation in which the voice user-interface receives an utterance from the user and presents a selection of the voice content to the user in response to the utterance.

13. The method of claim 12 , wherein the integrating is automatically performed.

14. The method of claim 13 , further comprising automatically associating a keyword with each topic which when uttered by the user cause the voice user-interface to play the topic associated with the uttered keyword.

15. The method of claim 14 , wherein if a keyword is associated with more than one topic, automatically generating a disambiguating prompt to played to the user to elicit an utterance of a keyword that is exclusively associated with one of the topics.

16. A computer readable medium having stored thereon a sequence of instructions, which when executed by a computer cause the computer to perform a method for building a voice response system, the method comprising:

defining an interaction between the voice response system and a user, the interaction comprising text elements to be played to the user and utterances to be spoken by the user, the text elements and the utterances being arranged in a form of a pseudo-dialog wherein the text elements to be played in response to an utterance are identified, said defining the interaction comprising automatically generating at least some of the text elements by automatically generating topic elements from text of a natural language text document by partitioning the text document into topics based on a frequency of occurrence of nouns therein, wherein each topic element comprises a subsection of the text document that relates to a particular topic; and

automatically generating code that uses the interaction to create a voice user-interface.

17. A computer readable medium having stored thereon a sequence of instructions which when executed by a computer cause the computer to perform a method for building a voice response system, the method comprising:

developing voice content for the voice response system, the voice content including prompts and information to be played to a user, said developing comprising reading a natural language text document and automatically generating topics therefrom corresponding to the information to be played to the user based on a frequency of occurrence of nouns in the text document; and

integrating the voice content with logic to define a voice user-interface that is capable of interacting with the user in a manner of a conversation in which the voice user-interface receives an utterance from the user and presents a selection of the voice content to the user in response to the utterance.

18. The computer readable medium of claim 17 , wherein the integrating is automatically performed.

19. A voice response system, comprising:

a processor; and

a memory coupled to the processor, the memory storing instructions which when executed by the processor cause the voice response system to perform a method comprising:

defining an interaction between the voice response system and a user, the interaction comprising text elements to be played to the user and utterances to be spoken by the user, the text elements and the utterances being arranged in a form of a pseudo-dialog wherein the text elements to be played in response to an utterance are identified, said defining the interaction comprising automatically generating at least some of the text elements by automatically generating topic elements from text of a natural language text document by partitioning the text document into topics based on a frequency of occurrence of nouns therein, wherein each topic element comprises a subsection of the text document that relates to a particular topic; and

automatically generating code that uses the interaction to create a voice user-interface.

20. The system of claim 19 , wherein the method further comprises generating keywords for each topic based on a frequency of occurrence of words in the text of a topic, the code then comprising instructions to select that topic to be played which has a keyword that matches an utterance of the user.

21. A voice response system, comprising:

a processor; and

a memory coupled to the processor, the memory starting instructions which when executed by the processor cause the voice response system to perform a method comprising:

developing voice content for the voice response system, the voice content including prompts and information to be played to a user, said developing comprising reading a natural language text document and automatically generating topics therefrom corresponding to the information to be played to the user based on a frequency of occurrence of nouns in the text document; and

integrating the voice content with logic to define a voice user-interface that is capable of interacting with the user in a manner of a conversation in which the voice user-interface receives an utterance from the user and presents a selection of the voice content to the user in response to the utterance.

22. The system of claim 21 , wherein the integrating is automatically performed.