Method and apparatus for creating modifiable and combinable speech objects for acquiring information from a speaker in an interactive voice response system

• US 6,314,402 B1
• Filed: 04/23/1999
• Issued: 11/06/2001
• Est. Priority Date: 04/23/1999
• Status: Expired due to Term

• Current Assignee: Nuance Communications Corporation of California
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Sinai, Julian, Gupta, Sundeep, Ehrlich, Steven C., Thirumalai, Madhavan, Klenk, Mark, Ghosh, Debajit, Monaco, Peter C.
• Primary Examiner(s): Dorvil, Richemond
• Assistant Examiner(s): MCFADDEN, SUSAN IRIS
• Application Number:
  US09/296,191
• Time in Patent Office: 928 Days
• Field of Search: 704/270, 704/272, 704/278
• US Class Current: 704/275
• CPC Class Codes: H04M 3/493 Interactive information services, e.g. directory enquiries ; Arrangements therefor, e.g. interactive voice response [IVR] systems or voice portals

Abstract

A method and apparatus for creating modifiable and combinable Speech Objects for use in an interactive voice response (IVR) environment. Each Speech Object is for acquiring a particular type of information from a speaker during an interaction between the speaker and a speech recognition mechanism. A Speech Object is an instantiation of a user-extensible class that includes properties, such as prompts and grammars, associated with the corresponding type of interaction. A Speech Object further includes logic for controlling the interaction with the user when executed in a processing system. A Speech Object can be subclassed to add additional properties and functionality to create customized Speech Objects, or such properties can be altered at runtime. Multiple Speech Objects, each for acquiring a particular type of information, can be combined to form a compound Speech Object.

Claims

1. A method of creating a device for defining a dialog interaction between a speaker and a speech recognition mechanism, the method comprising:

providing a set of properties associated with the dialog interaction and logic for using the set of properties to control the dialog interaction when executed in a processing system; and

defining a class to include the set of properties and the logic, such that the class can be instantiated in the processing system as an object configured to control the dialog interaction, such that one or more sub-classes are definable based on the class, each subclass including properties inherited from the class, and such that the object can control the dialog interaction without the need for a dialog manager external to said object, wherein the object implements a predefined interface, such that the object and a plurality of other similarly constructed objects for controlling dialog interaction can be invoked using the predefined interface.

2. A method as recited in claim 1, wherein the object is configured to package results of the interaction into a playable object, such that when invoked, said playable object causes audio data representing the result to be played via an audio interface.

3. A method as recited in claim 1, wherein any subclass of said one or more subclasses may be defined as a specialization of said class.

4. A method as recited in claim 1, further comprising defining a subclass of the class, such that the subclass can be instantiated as an object in the processing system, including defining the subclass to include a second set of properties, the second set of properties including at least some of said set of properties and additional properties not part of said set of properties.

5. A method as recited in claim 1, wherein the set of properties comprises a set of prompts associated with the interaction.

6. A method as recited in claim 1, wherein the set of properties comprises a set of grammars associated with the interaction.

7. A method as recited in claim 1, wherein said defining comprises defining the class such that the object installs the set of grammars dynamically when invoked in the processing system.

8. A method as recited in claim 1, wherein the set of properties comprises a set of prompts and a set of grammars, each associated with the interaction.

9. A machine-readable storage medium having stored therein information for configuring an interactive voice response platform to perform an interaction with a speaker, the information comprising:

information representing a set of properties associated with the interaction;

logic for using the set of properties to control the interaction when the logic is executed in a processing system; and

information defining the set of properties and the logic to be elements of a class that can be instantiated as a speech object in the processing system to control the interaction, such that one or more sub-classes are definable based on the class, each subclass including properties inherited from the class, and such that the speech object can control the dialog interaction without the need for a dialog manager external to the speech object, wherein the speech object implements a predefined interface by which the speech object and a plurality of other speech objects can be invoked.

10. A machine-readable storage medium as recited in claim 9, wherein said information representing the set of properties associated with the interaction comprises:

information representing a set of prompts associated with the interaction; and

information representing a set of grammars associated with the interaction.

11. A machine-readable storage medium having stored therein information readable by a processing system, the information comprising information defining a class that can be instantiated as an object in a processing system to control a dialog interaction between a speaker and a speech recognition mechanism, the class having a set of properties associated with the dialog interaction and logic for using the set of properties to control the dialog interaction when the logic is executed, such that one or more sub-classes are definable based on the class, each subclass including properties inherited from the class, and such that the object can control the dialog interaction without the need for a dialog manager external to said object, wherein the object implements a predefined interface using which the object and a plurality of other similarly constructed objects for controlling dialog interaction can be invoked.

12. A method as recited in claim 11, wherein the set of properties comprises a set of prompts, a set of grammars, or both, each associated with the dialog interaction.

13. A method of creating a software component for defining interaction between a speaker and a speech recognition mechanism in an interactive voice response environment, the method comprising:

including first data in the software component, the first data representing a set of prompts that can be output to the speaker when the software component is invoked by a processing system;

including second data in the software component, the second data representing a set of grammars associated with the interaction;

including first code in the software component, the first code representing processing logic for controlling the interaction when executed by the processing system, based on the set of prompts and the set of grammars; and

including second code in the software component, the second code for defining the software component as a first class that can be instantiated by the processing system as an object for controlling the interaction between the speaker and the speech recognition mechanism, such that the first class is extensible by definition of one or more subclasses of the first class, each said subclass inheriting properties of the first class, such that the object implements a predefined interface by which the object and a plurality of other similarly constructed objects can be invoked;

wherein the object can control the interaction without the need for a dialog manager external to said object.

14. A method as recited in claim 13, wherein said including second code comprises including the second code such that the first class can be combined with a second class to form a third class separate from the first class and the second class, such that the third class can be instantiated by the processing system as one or more objects.

15. A method of creating a device for defining an interaction between a speaker and a speech recognition mechanism, the method comprising:

providing information representing a first class in an interactive voice response environment, such that the first class can be instantiated as one or more objects in a processing system to control the interaction, such that one or more sub-classes are definable based on the class, each subclass including properties inherited from the class, and such that the object can control the dialog interaction without the need for a dialog manager external to said object, wherein the object implements a predefined interface by which the object and a plurality of other similarly constructed objects can be invoked; and

using a computer system to define a second class as a specialization of the first class, the second class including a set of prompts associated with the interaction, a set of grammars associated with the interaction, and logic for using the set of prompts and the set of grammars when executed on the processing system to control the interaction between the speaker and the speech recognition mechanism, such that the second class can be instantiated as an object in the processing system to control the interaction.

16. A method as recited in claim 15, wherein said using a computer system to define the second class comprises defining the second class as a subclass of the first class.

17. A method as recited in claim 16, wherein the first class includes a first set of prompts and a first set of grammars, and wherein said using a computer system to define the second class further comprises:

defining the second class to include a second set of prompts, the second set of prompts including at least one prompt of said first set of prompts and a prompt that is not part of said set of prompts; and

defining the second class to include a second set of grammars, the second set of grammars including at least one grammar of said first set of grammars and a grammar that is not part of said set of grammars.

18. A method of creating a compound device for defining an interaction between a speaker and a speech recognition mechanism, the method comprising:

selecting a plurality of classes, each of the plurality of classes defining operations for an interaction of a particular type between a speaker and a speech recognition mechanism in an interactive voice response environment, each of the plurality of classes having associated with it a set of prompts, a set of grammars, or both, and logic for using the set of prompts, the set of grammars, or both, to control an interaction between the speaker and the speech recognition mechanism when executed on a processing system, such that each of the plurality of classes can be instantiated as a speech object configured to control an interaction of the corresponding type, such that one or more sub-classes are definable based on each of the classes, each subclass including properties inherited from the corresponding class, and such that each speech object can control the dialog interaction without the need for a dialog manager external to said speech object, wherein each speech object implements an interface common to all of the speech objects; and

using a computer system to define a compound speech object class for use in the interactive voice response environment, such that the compound speech object class, when instantiated in a processing system as a compound speech object, encapsulates the speech objects representing said selected plurality of classes, the compound speech object having logic for executing the speech objects in a specified order during the interaction with the speaker.

19. A method as recited in claim 18, further comprising using the computer system to define the compound class, such that the compound speech object further encapsulates a processing object separate from the plurality of speech objects, the processing object providing processing logic.

20. A method as recited in claim 19, further comprising using the computer system to define the compound class such that a first one of the plurality of speech objects encapsulated in said compound speech object encapsulates a plurality of additional speech objects, such that said first one of the plurality of speech objects is also a compound speech object.

21. An interactive voice response (IVR) system comprising:

a speech recognition unit;

an audio interface configured to communicate audio information with a speaker; and

an IVR platform coupled to the speech recognition unit and to the audio interface, the IVR platform including

a speech-enabled application; and

a speech object invocable in response to the application to control a particular type of audio interaction with the speaker, wherein the speech object is an instantiation of a class, the class having a set of properties associated with a corresponding type of interaction and logic for using the set of properties to control an interaction of said type when the logic is executed, such that one or more sub-classes are definable by a user based on the class, each subclass including properties inherited from the class, and such that the speech object controls the interaction with the speaker without the need for a dialog manager external to the speech object, wherein the speech object implements a predefined interface using which a plurality of similarly constructed speech objects can be invoked.

22. An IVR system as recited in claim 21, wherein each said subclass is a customized speech object including properties inherited from said class.

23. An IVR system as recited in claim 21, wherein the set of properties associated with the interaction comprises a set of prompts associated with the interaction.

24. An IVR system as recited in claim 21, wherein the set of properties associated with the interaction comprises a set of grammars associated with the interaction.

25. An IVR system as recited in claim 21, further comprising a speech channel object providing the IVR with access to the audio interface and the speech recognition unit, wherein the speech channel object is an instantiation of a speech channel class.

26. An IVR system as recited in claim 21, wherein the audio interface comprises a telephony interface.

27. An IVR system as recited in claim 26, wherein the audio interface comprises an Internet Protocol (IP) based interface.

28. An interactive voice response (IVR) system comprising:

interface means for communicating audio information with a speaker;

recognition means for performing speech recognition on a portion of the audio information that is received from the speaker;

means for executing a speech-enabled application, including means for requesting an interaction with the speaker to acquire said portion of the audio information; and

means for invoking a speech object to control the interaction, wherein the speech object is an instantiation of a class, the class having a set of properties associated with the interaction and logic for using the set of properties to control the interaction when the logic is executed, such that one or more sub-classes can be defined based on the class, each subclass including properties inherited from the class, such that the speech object can control the dialog interaction without the need for a dialog manager external to the speech object, wherein the speech object implements a predefined interface by which the speech object and a plurality of other speech objects can be invoked.

29. An IVR system as recited in claim 28, wherein each said subclass represents a customized speech object including properties inherited from the class.

30. An IVR system as recited in claim 28, wherein the set of properties associated with the interaction comprises a set of prompts associated with the interaction.

31. An IVR system as recited in claim 28, wherein the set of properties associated with the interaction comprises a set of grammars associated with the interaction.

32. An IVR system as recited in claim 28, further comprising means for providing the IVR with access to the interface means and the recognition means by invoking a speech channel object as an instantiation of a speech channel class.

33. An IVR system as recited in claim 28, wherein the audio interface comprises a telephony interface.

34. An IVR system as recited in claim 33, wherein the audio interface comprises an Internet Protocol (IP) based interface.

35. A device for configuring a processing system for acquisition of information from a speaker in an interactive voice response (IVR) environment, the device comprising:

a machine-readable storage medium; and

information stored in the machine-readable storage medium, the information defining a class for use in the IVR environment, such that the class can be instantiated in the IVR environment as a compound object encapsulating a plurality of objects, each of the plurality of objects for configuring the IVR environment to acquire a particular type of information from the speaker during an interaction with the speaker, each of the plurality of objects invocable in a specified order during the interaction, wherein one or more sub-classes are definable based on the class, each subclass including properties inherited from the class, wherein the compound object can control the dialog interaction without the need for a dialog manager external to the compound object, and wherein the compound object implements a predefined interface by which the compound object and a plurality of other objects can be invoked.

36. A device as recited in claim 35, wherein the information comprises, for each of the plurality of objects:

information representing a set of properties associated with the interaction;

logic for using the set of properties to control the interaction when the logic is executed in a processing system; and

information defining the set of properties and the logic to be elements of a user-extensible class.

37. A device as recited in claim 36, wherein said information representing the set of properties associated with the interaction comprises:

information representing a set of prompts associated with the interaction; and

information representing a set of grammars associated with the interaction.

38. A device as recited in claim 35, wherein said stored information is such that the compound speech object further encapsulates a processing object separate from the plurality of objects, the processing object having processing logic.

39. A device as recited in claim 35, wherein said stored information is such that a first one of the plurality of objects encapsulated in the compound speech object encapsulates a plurality of additional speech objects.