Markup language-based selection and utilization of recognizers for utterance processing

• US 20100185447A1
• Filed: 01/22/2009
• Published: 07/22/2010
• Est. Priority Date: 01/22/2009
• Status: Active Grant

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Krumel, Andrew K., Masse, Pierre-Alexandre F., Ruff, Joseph A.
• Granted Patent: US 8,515,762 B2
• Time in Patent Office: Days
• US Class Current: 704/255
• CPC Class Codes: G10L 15/32 Multiple recognisers used in sequence or in parallel; Score combination systems therefor, e.g. voting systems

Abstract

Embodiments are provided for selecting and utilizing multiple recognizers to process an utterance based on a markup language document. The markup language document and an utterance are received in a computing device. One or more recognizers are selected from among the multiple recognizers for returning a results set for the utterance based on markup language in the markup language document. The results set is received from the one or more selected recognizers in a format determined by a processing method specified in the markup language document. An event is then executed on the computing device in response to receiving the results set.

Claims

1. A computer-implemented method for utilizing a plurality of recognizers to process an utterance based on a markup language document, comprising:

receiving the markup language document;

receiving the utterance;

selecting at least one of the plurality of recognizers for returning a results set for the utterance based on markup language in the markup language document;

receiving the results set from the selected at least one of the plurality of recognizers in a format determined by a processing method specified in the markup language document; and

executing an event in response to receiving the results set.

2. The method of claim 1 further comprising:

sending version data for the at least one of the plurality of recognizers from a first computing device to a second computing device; and

if the version data identifies an updated version for the at least one of the plurality of recognizers, then receiving additional markup language from the second computing device at the first computing device, the additional markup language comprising instructions for utilizing newly added features in the updated version.

3. The method of claim 1, wherein receiving the markup language document comprises at least one of receiving a locally stored markup language document on a client computing device and receiving a remotely stored markup language document from a network server on the client computing device.

4. The method of claim 1, wherein selecting at least one of the plurality of recognizers for returning a result for the utterance based on markup language in the markup language document comprises:

recognizing a grammar used in the utterance;

parsing the markup language document for at least one markup language tag identifying at least one of the plurality of recognizers for returning the results set for the utterance based on the grammar; and

selecting the at least one of the plurality of recognizers identified by the at least one markup language tag.

5. The method of claim 4 further comprising at least one of sending and streaming the utterance to the selected at least one of the plurality of recognizers.

6. The method of claim 4, wherein selecting the at least one of the plurality of recognizers identified by the at least one markup language tag comprises at least one of selecting a local recognizer embedded on a client computing device and selecting a remote recognizer stored on a network server.

7. The method of claim 1, wherein receiving the results set from the selected at least one of the plurality of recognizers in a format determined by a processing method specified in the markup language document comprises receiving the results set in at least one of an unprocessed results format and a processed results format, wherein the unprocessed results format comprises one or more terms comprising a speech recognition of the utterance and wherein the processed results format comprises a speech recognition of one or more terms associated with the utterance.

8. The method of claim 1, wherein executing an event in response to receiving the results set comprises determining an action in response to receiving the results set, wherein the action is based on at least one of:

a number of results in the results set, a confidence score based on the results set, a return code for the results set, and predetermined values returned in the results set, and wherein the action comprises at least one of;

displaying the results set to a user, performing a voice-activated dialing function based on the results set, ignoring the results set, and combining the results set with at least one previously received results set to create a plurality of results sets.

9. A system for utilizing a plurality of recognizers to process an utterance based on a markup language document, the system comprising a client computing device, the client computing device comprising:

a memory for storing executable program code; and

a processor, functionally coupled to the memory, the processor being responsive to computer-executable instructions contained in the program code and operative;

to receive the markup language document;

to receive the utterance;

to select at least one of the plurality of recognizers for returning a results set for the utterance based on markup language in the markup language document;

to receive the results set from the selected at least one of the plurality of recognizers in a format determined by a processing method specified in the markup language document; and

to execute an event in response to receiving the results set.

10. The system of claim 9, wherein the processor is further operative to:

determine whether an updated version is available for at least one of the plurality of recognizers; and

if an updated version is available for the at least one of the plurality of recognizers, then receive an updated markup language document comprising additional markup language for handling newly added features in the updated version.

11. The system of claim 9, wherein the markup language document comprises at least one of a locally stored markup language document on the client computing device and a remotely stored markup language document on a network server.

12. The system of claim 9, wherein the processor, in selecting at least one of the plurality of recognizers for returning a results set for the utterance based on markup language in the markup language document, is further operative to:

recognize a grammar used in the utterance;

parse the markup language document for at least one markup language tag identifying at least one of the plurality of recognizers for returning the results set for the utterance based on the grammar; and

select the at least one of the plurality of recognizers identified by the at least one markup language tag.

13. The system of claim 12, wherein the processor is further operative to send the utterance to the selected at least one of the plurality of recognizers.

14. The system of claim 12, the selected at least one of the plurality of recognizers identified by the at least one markup language tag comprises at least one of a local recognizer embedded on the client computing device and a remote recognizer stored on a network server.

15. The system of claim 9, wherein the results set from the selected at least one of the plurality of recognizers is received in a format determined by a processing method specified in the markup language document, wherein the format comprises at least one of an unprocessed results format and a processed results format, wherein the unprocessed results format comprises one or more terms comprising a speech recognition of the utterance and wherein the processed results format comprises one or more terms associated with a speech recognition of the utterance.

16. The system of claim 9, wherein the processor, in executing an event in response to receiving the results set, is further operative to determine an action in response to receiving the results set, wherein the action is based on at least one of:

a number of results in the results set, a confidence score based on the results set, a return code for the results set, and predetermined values returned in the results set, and wherein the action comprises at least one of;

displaying the results set to a user, performing a voice-activated dialing function based on the results set, ignoring the results set, and combining the results set with at least one previously received results set to create a plurality of results sets.

17. A computer-readable storage medium comprising computer executable instructions which, when executed on a computer, will cause the computer to perform a method for utilizing a plurality of recognizers to process an utterance based on a markup language document, comprising:

receiving the markup language document, wherein receiving the markup language document comprises at least one of receiving a locally stored markup language document on a client computing device and receiving a remotely stored markup language document from a network server on the client computing device.

receiving the utterance;

selecting at least one of the plurality of recognizers for returning a results set for the utterance based on markup language in the markup language document, wherein selecting at least one of the plurality of recognizers for returning a results set for the utterance based on markup language in the markup language document comprises;

recognizing a grammar used in the utterance;

parsing the markup language document for at least one markup language tag identifying at least one of the plurality of recognizers for returning the results set for the utterance based on the grammar; and

selecting the at least one of the plurality of recognizers identified by the at least one markup language tag;

streaming the utterance to the selected at least one of the plurality of recognizers;

receiving the results set from the selected at least one of the plurality of recognizers in a format determined by a processing method specified in the markup language document; and

executing an event in response to receiving the results set.

18. The computer-readable storage medium of claim 17 further comprising:

determining whether an updated version is available for at least one of the plurality of recognizers; and

if an updated version is available for the at least one of the plurality of recognizers, then receiving an updated markup language document comprising additional markup language for handling newly added features in the updated version.

19. The computer-readable storage medium of claim 17, wherein receiving the results set from the selected at least one of the plurality of recognizers in a format determined by a processing method specified in the markup language document comprises receiving the results set in at least one of an unprocessed results format and a processed results format, wherein the unprocessed results format comprises one or more terms comprising a speech recognition of the utterance and wherein the processed results format comprises a speech recognition of one or more terms associated with the utterance.

20. The computer-readable storage medium of claim 17, wherein executing an event in response to receiving the results set determining an action in response to receiving the results set, wherein the action is based on at least one of:

a number of results in the results set, a confidence score based on the results set, a return code for the results set, and predetermined values returned in the results set, and wherein the action comprises at least one of;

displaying the results set to a user, performing a voice-activated dialing function based on the results set, ignoring the results set, and combining the results set with at least one previously received results set to create a plurality of results sets.