Method and system to parameterize dialog systems for the purpose of branding

• US 7,877,257 B2
• Filed: 09/28/2005
• Issued: 01/25/2011
• Est. Priority Date: 09/27/2004
• Status: Active Grant

• Current Assignee: Robert Bosch Corporation (Robert Bosch GmbH), Volkswagen Of America (Volkswagen AG)
• Original Assignee: Robert Bosch Corporation (Robert Bosch GmbH), Volkswagen Of America (Volkswagen AG)
• Inventors: Mishra, Rohit, Raghunathan, Badri, Schmidt, Hauke, Bergmann, Carsten, Weng, Fuliang
• Primary Examiner(s): Vo; Huyen X.
• Application Number:
  US11/238,511
• Publication Number:
  US 20060100875A1
• Time in Patent Office: 1,945 Days
• Field of Search: 704/251, 704/257, 704/231, 704/258, 704/7, 704/9, 704/270, 704/275, 704/270.1, 704/274, 704/272, 340/425.5
• US Class Current: 704/257
• CPC Class Codes: G06F 40/295 Named entity recognition
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/227 of the speaker; Human-factor methodology

Abstract

A method and system to parameterize a spoken language dialog system, includes providing a storage area to store at least one parameter value, integrating a subcomponent with at least one module of the spoken language dialog system, and configuring the subcomponent to access the at least one parameter value and to adjust an operation or output of the module based on the at least one parameter value.

Claims

1. A computer-implemented method to parameterize a spoken language dialog system, the spoken language dialog system including a speech recognition module, a natural language understanding module, a dialog management module, a response generation module, and a text-to-speech module, the method comprising:

detecting, by a computer processor, at least one characteristic of a speech input of a speaker using at least one of the speech recognition module and the natural language understanding module;

setting, by the computer processor, at least one parameter value in a storage area in accordance with the detected at least one characteristic; and

executing, by the computer processor, a subcomponent of one of the modules to subsequently access the set at least one parameter value and to adjust output of the spoken language dialog system based on the at least one parameter value;

wherein the adjustment includes adjusting a stored intrusiveness level parameter of the dialogue system, and wherein, based on a comparison of an importance level of an external event and the intrusiveness level parameter of the dialogue system, the spoken language dialog system determines whether or not to initiate a dialog that informs the speaker of the external event.

2. The method of claim 1, wherein a subcomponent of the speech recognition module adjusts a conversion of a digitized acoustic sample into at least one of a sequence of words and a directed graph generated by the speech recognition module.

3. The method of claim 1, wherein a subcomponent of the natural language understanding module adjusts an analysis of at least one of a sequence of words and directed group of word hypothesis for syntactic and semantic structure.

4. The method of claim 3, wherein the subcomponent of the natural language understanding module is configured to adjust the analysis of at least one of a sequence of words and directed group of word hypothesis for syntactic and semantic structure with respect to a user speaking style.

5. The method of claim 1, wherein a subcomponent of the dialog management module adjusts at least one of an assembly of a command for a device and creation of a response description.

6. The method of claim 5, wherein the subcomponent of the dialog management module is configured to adjust the at least one of the assembly of the command for the device and the creation of the response description with respect to a dialog style of the spoken dialog system.

7. The method of claim 6, wherein the dialog style includes a relative response length.

8. The method of claim 6, wherein the subcomponent of the dialog management module is configured to adjust the at least one of the assembly of the command for the device and the creation of the response description with respect to the intrusiveness level of the spoken dialog system.

9. The method of claim 1, wherein a subcomponent of the response generation module adjusts a translation of a response description into a sequence of at least one of words and acoustic icons.

10. The method of claim 9, wherein the subcomponent of the response generation module is configured to adjust the translation of the response description into the sequence of at least one of the words and acoustic icons with respect to an amount of response to be provided to a user of the spoken language dialog system.

11. The method of claim 9, wherein the subcomponent of the response generation module is configured to adjust the translation of the response description into the sequence of the at least one of the words and acoustic icons with respect to a choice of the at least one of the words and acoustic icons.

12. The method of claim 1, wherein a subcomponent of the text-to-speech module adjusts a conversion of a sequence of at least one of words into digitalized acoustic samples.

13. The method of claim 12, wherein the subcomponent of the text-to-speech module is configured to adjust the conversion of the sequence of words into digitalized acoustic samples with respect to prosody control.

14. The method of claim 12, wherein the subcomponent of the text-to-speech module is configured to adjust the conversion of the sequence of words into digitalized acoustic samples with respect to a voice characteristic.

15. The method of claim 14, further comprising:

providing an interface to set at least one of the at least one parameter and the at least one profile.

16. The method of claim 1, further comprising:

organizing the at least one parameter into at least one profile.

17. The method of claim 1, wherein the external event is one of a warning of an impending accident and an indication of a low fuel level of a vehicle.

18. The method of claim 1, further comprising:

in accordance with the detection, categorizing the speaker as belonging to one of a plurality of speaker categories;

wherein;

the storage area stores a plurality of parameter values;

each of the plurality of parameter values is associated with a respective one of the plurality of speaker categories; and

the at least one parameter is a subset of the plurality of parameter values that correspond to the speaker category to which the speaker has been categorized as belonging, the access of the at least one parameter by the subcomponent being based on the categorization of the speaker.

19. The method of claim 1, wherein the device is a motor vehicle.

20. The method of claim 19, wherein the spoken language dialog system is on board the motor vehicle.

21. The method of claim 20, wherein different settings of the at least one parameter are provided to a user for user-selection of one of the different settings.

22. The method of claim 1, wherein the intrusiveness level indicates a threshold importance required for initiating the dialog that informs the speaker of the external event.

23. A hardware arrangement configured to parameterize a spoken language dialog system, the spoken language dialog system including a speech recognition module, a natural language understanding module, a dialog management module, a response generation module, and a text-to-speech module, the arrangement comprising:

a storage area to store at least one parameter value; and

a computer processor;

wherein;

the computer processor is configured to;

detect at least one characteristic of a speech input of a speaker using at least one of the speech recognition module and the natural language understanding module;

set the at least one parameter value in the storage area in accordance with the detected at least one characteristic; and

execute a subcomponent integrated with one of the modules to, subsequent to the setting of the at least one parameter value, access the set at least one parameter value and to adjust output of the spoken language dialog management system based on the accessed at least one parameter value; and

the adjustment of the output includes adjusting a stored intrusiveness level parameter of the dialogue system, and wherein, based on a comparison of an importance level of an external event and the intrusiveness level parameter of the dialogue system, the spoken language dialog system determines whether or not to initiate a dialog that informs the speaker of the external event.

24. The arrangement of claim 23, wherein a subcomponent of the speech recognition module adjusts a conversion of a digitized acoustic sample into at least one of a sequence of words and a directed graph generated by the speech recognition module.

25. The arrangement of claim 23, wherein a subcomponent of the natural language understanding module adjusts an analysis of at least one of a sequence of words and directed group of word hypothesis for syntactic and semantic structure.

26. The arrangement of claim 25, wherein the subcomponent of the natural language understanding module is configured to adjust the analysis of at least one of a sequence of words and directed group of word hypothesis for syntactic and semantic structure with respect to a user speaking style.

27. The arrangement of claim 23, wherein a subcomponent of the dialog management module adjusts at least one of an assembly of a command for a device and creation of a response description.

28. The arrangement of claim 27, wherein the subcomponent of the dialog management module is configured to adjust the at least one of the assembly of the command for the device and the creation of the response description with respect to a dialog style of the spoken dialog system.

29. The arrangement of claim 28, wherein the dialog style includes a relative response length.

30. The arrangement of claim 27, wherein the subcomponent of the dialog management module is configured to adjust the at least one of the assembly of the command for the device and the creation of the response description with respect to the intrusiveness level of the spoken dialog system.

31. The arrangement of claim 23, wherein a subcomponent of the response generation module adjusts a translation of a response description into a sequence of at least one of words and acoustic icons.

32. The arrangement of claim 31, wherein the subcomponent of the response generation module is configured to adjust the translation of the response description into the sequence of at least one of the words and acoustic icons with respect to an amount of response to be provide to a user of the spoken dialog system.

33. The arrangement of claim 31, wherein the subcomponent of the response generation module is configured to adjust the translation of the response description into the sequence of the at least one of the words and acoustic icons with respect to a choice of the at least one of the words and acoustic icons.

34. The arrangement of claim 23, wherein a subcomponent of the text-to-speech module adjusts a conversion of a sequence of at least one of words into digitalized acoustic samples.

35. The arrangement of claim 34, wherein the subcomponent of the text-to-speech module is configured to adjust the conversion of the sequence of words into digitalized acoustic samples with respect to prosody control.

36. The arrangement of claim 34, wherein the subcomponent of the text-to-speech module is configured to adjust the conversion of the sequence of words into digitalized acoustic samples with respect to a voice characteristic.

37. A spoken language dialog system, comprising:

a storage area to store at least one parameter value; and

a computer processor configured to execute;

a plurality of modules to perform subtasks of the spoken language dialog system, the plurality of modules including a speech recognition module, a natural language understanding module, a dialog management module, a response generation module, and a text-to-speech module; and

a subcomponent integrated with the one of the modules to access the at least one parameter value and to adjust an operation or output of the spoken language dialog system based on the at least one parameter value;

wherein;

at least one characteristic of a speech input of a speaker is detected using at least one of the speech recognition module and the natural language model;

the at least one parameter value is set in storage in accordance with the detected characteristic;

the computer processor is configured to execute the subcomponent to, subsequent to the setting of the at least one parameter value, access the set at least one parameter value and to adjust output of the spoken language dialog system based on the accessed at least one parameter value; and

the adjustment of the output includes adjusting a stored intrusiveness level parameter of the dialogue system, and wherein, based on a comparison of an importance level of an external event and the intrusiveness level parameter of the dialogue system, the spoken language dialog system determines whether or not to initiate a dialog that informs the speaker of the external event.

38. The spoken language dialog system of claim 37, wherein the natural language understanding module is configured to perform an analysis of at least one of a sequence of words and directed group of word hypothesis for syntactic and semantic structure, and a subcomponent of the natural language understanding module is configured to adjust the analysis with respect to a user speaking style.

39. The spoken language dialog system of claim 37, wherein the dialog management module is configured to at least one of assemble a command for a device and create of a response description, and a subcomponent of the dialog management module is configured to adjust the at least one of the assembly of the command for the device and the creation of the response description with respect to one of a dialog style and intrusiveness of the spoken dialog system.

40. The spoken language dialog system of claim 39, wherein the dialog style includes a relative response length.

41. The spoken language dialog system of claim 37, wherein the response generation module is configured to perform a translation of a response description into a sequence of at least one of words and acoustic icons, and a subcomponent of the response generation module is configured to adjust the translation with respect to one of an amount of response to be provided to a user of the spoken dialog system and a choice of the at least one of the words and acoustic icons.

42. The spoken language dialog system of claim 37, wherein the text-to-speech module is configured to perform a conversion of a sequence of at least one of words into digitalized acoustic samples, and a subcomponent of the text-to-speech module is configured to adjust the conversion with respect to at least one of prosody control and a voice characteristic.

43. A computer-implemented method to parameterize a spoken language dialog system, the spoken language dialog system including a speech recognition module, a natural language understanding module, a dialog management module, a response generation module, and a text-to-speech module, the method comprising:

detecting, by a computer processor, at least one characteristic of a speech input of a speaker using at least one of the speech recognition module and the natural language module;

setting, by the computer processor, at least one parameter value in a storage area in accordance with the detected at least one characteristic; and

executing, by the computer processor, a subcomponent of the spoken language dialog system to subsequently access the set at least one parameter value and adjust output of the spoken language dialog system based on the at least one parameter value;

wherein the adjustment includes adjusting a stored intrusiveness level parameter of the dialogue system, and wherein, based on a comparison of an importance level of an external event and the intrusiveness level parameter of the dialogue system, the spoken language dialog system determines whether or not to initiate a dialog that informs the speaker of the external event.

44. A computer-implemented method to parameterize a spoken language dialog system, the spoken language dialog system including a speech recognition module, a natural language understanding module, a dialog management module, a response generation module, and a text-to-speech module, the method comprising:

detecting, by a computer processor, at least one characteristic of a speech input of a speaker using at least one of the speech recognition module and the natural language understanding module;

setting, by the computer processor, at least one parameter value in a storage area in accordance with the detected at least one characteristic;

executing, by the computer processor, a subcomponent of one of the modules to subsequently access the set at least one parameter value;

adjusting a stored intrusiveness level parameter of the dialogue system; and

for each of a plurality of external events;

comparing the intrusiveness level parameter of the dialogue system with a predefined importance level of the respective external event; and

determining whether or not to initiate a dialog that informs the speaker of the respective external event based on the comparison.