Automated speech recognition proxy system for natural language understanding

• US 8,484,031 B1
• Filed: 03/24/2011
• Issued: 07/09/2013
• Est. Priority Date: 01/05/2011
• Status: Active Grant

• Current Assignee: Interactions, LLC
• Original Assignee: Interactions Corporation (Interactions, LLC)
• Inventors: Yeracaris, Yoryos, Gray, Philip M., Dreher, James P.
• Primary Examiner(s): MCFADDEN, SUSAN IRIS
• Application Number:
  US13/070,865
• Time in Patent Office: 838 Days
• Field of Search: 704/270.1
• US Class Current: 704/270.1
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications

Abstract

An interactive response system mixes HSR subsystems with ASR subsystems to facilitate overall capability of voice user interfaces. The system permits imperfect ASR subsystems to nonetheless relieve burden on HSR subsystems. An ASR proxy is used to implement an IVR system, and the proxy decides, based on a set of rules, to route an utterance solely to one ASR, to route it to an HSR in addition to at least one ASR, to route it solely to one or more HSR subsystem, to reroute an utterance originally sent to an ASR over to an HSR, to use HSRs to help tune and train one or more ASRs, and to use multiple ASRs to increase reliability of results.

Claims

1. A computer-implemented system for processing an interaction, the interaction including an utterance requiring recognition before being usable for further computer-implemented processing, the system comprising:

an application configured to provide the utterance;

a proxy processing subsystem in communication with the application, the proxy processing subsystem configured to receive the utterance and initiate processing thereof;

a recognition decision engine configured to receive from the proxy processing system an utterance for recognition and ancillary information regarding the utterance for recognition, the recognition decision engine selecting, responsive to the ancillary information, one or more recognizers from a first type of recognizer subsystems and a second type of recognizer subsystems; and

a results decision engine operably coupled with the one or more recognizers and configured to return to the proxy processing subsystem a recognition result, the results decision engine further automatically updating a statistics database responsive to results of processing by the one or more recognizers.

2. The system of claim 1, wherein the application is further configured to provide rules for processing the utterance and wherein the proxy processing subsystem is further configured to process the utterance in accordance with the rules.

3. The system of claim 1, wherein the ancillary information comprises recognition parameters provided by the application.

4. The system of claim 1, wherein the utterance is a computer-stored sound file corresponding to human speech.

5. The system of claim 1, wherein the first type of recognizer subsystems comprises automatic speech recognition processors.

6. The system of claim 1, wherein the one or more recognizers comprises a plurality of automatic speech recognition processors.

7. The system of claim 1, wherein the recognition decision engine communicates with the statistics database when selecting the one or more recognizers.

8. A computer-implemented method for processing an interaction comprising:

receiving data representing an utterance from a computer application;

using a proxy processor, automatically initiating processing of the data;

selecting, using ancillary information regarding the utterance, one or more recognizers from a first type of recognizer subsystems and a second type of recognizer subsystems;

returning a recognition result responsive to results of processing by the one or more recognizers; and

automatically updating a statistics database responsive to results of processing by the one or more recognizers.

9. The computer-implemented method of claim 8, wherein said receiving further comprises receiving rules from the computer application and wherein said processing of the data is performed in accordance with the rules.

10. The computer-implemented method of claim 8, wherein the ancillary information comprises recognition parameters provided by the computer application.

11. The computer-implemented method of claim 8, wherein the data comprises a computer-stored sound file corresponding to human speech.

12. The computer-implemented method of claim 8, wherein the first type of recognizer subsystems comprises automatic speech recognition processors.

13. The computer-implemented method of claim 8, wherein the one or more recognizers comprises a plurality of automatic speech recognition processors.

14. The computer-implemented method of claim 8, wherein said selecting further comprises obtaining statistical information from the statistical database.

15. A system for processing an interaction, the interaction including an utterance requiring recognition before being usable for further computer-implemented processing, the system comprising non-transitory computer storage media storing programming instructions executable by at least one processor to:

receive data representing an utterance from a computer application;

using a proxy processor, automatically initiate processing of the data;

select, using ancillary information regarding the utterance, one or more recognizers from a first type of recognizer subsystems and a second type of recognizer subsystems;

return a recognition result responsive to results of processing by the one or more recognizers; and

automatically update a statistics database responsive to results of processing by the one or more recognizers.

16. The system of claim 15, wherein the instructions to receive further comprise instructions to receive rules from the computer application and wherein said processing of the data is performed in accordance with the rules.

17. The system of claim 15, wherein the ancillary information comprises recognition parameters provided by the computer application.

18. The system of claim 15, wherein the data comprises a computer-stored sound file corresponding to human speech.

19. The system of claim 15, wherein the first type of recognizer subsystems comprises automatic speech recognition processors.

20. The system of claim 15, wherein the instructions to select further comprise instructions to obtain statistical information from the statistical database.