System and method for providing transcription services using a speech server in an interactive voice response system

• US 20070203708A1
• Filed: 02/28/2006
• Published: 08/30/2007
• Est. Priority Date: 02/28/2006
• Status: Abandoned Application

• Current Assignee: Intervoice Limited Partnership A Nevada Limited Partnership Composed of As Its SOLE Partner Intervoice GP Incorporated
• Original Assignee: Intervoice Limited Partnership
• Inventors: Cave, Ellis, Polcyn, Michael, Waln, Kenneth, Blaszczak, Bogdan
• Application Number:
  US11/364,353
• Publication Number:
  US 20070203708A1
• Time in Patent Office: Days
• US Class Current: 704/270.100
• CPC Class Codes: G10L 15/30 Distributed recognition, e.g. in client-server systems, for mobile phones or network applications
• CPC Class Codes: H04M 2201/40 using speech recognition
• CPC Class Codes: H04M 2201/60 Medium conversion
• CPC Class Codes: H04M 3/42221 Conversation recording systems at the subscriber's set H04M1/656
• CPC Class Codes: H04M 3/4936 Speech interaction details speech recognition per se G10L15/00
• CPC Class Codes: H04M 3/5166 in combination with interactive voice response systems or voice portals, e.g. as front-ends

Abstract

The present invention is directed to a system and method in which an interface (proxy) is positioned between a browser and a speech server such that the proxy, while transparent to both the browser and the speech server, collects and stores data, including utterances and other media obtained from a user, such that the media data can be retrieved in a uniform manner for subsequent manipulation, such as, for example, transcription or presentation (or preservation) of a tangible format of the media as a function of a transaction session with the user. The proxy is a passive monitoring device positioned between the functioning components of a system such that the proxy looks to the browser as a speech server and looks to the speech server as a browser.

Claims

1. A system for capturing data flowing among a plurality of individual components of an interactive voice response (IVR) system, said system comprising:

a proxy for interposing between a browser and a speech server;

said proxy comprising;

a first interface for accepting communications from said browser, said communications sent from said browser in response to scripts, said communications containing both a control protocol and a voice stream, said control protocol understood by said speech server, said communications containing additional information not part of said protocol;

a second interface for delivering accepted ones of said communications to said speech server, and for accepting from said speech server communications for delivery to said browser, said communications from said speech server using a protocol understood by said browser; and

means for storing both said additional information and said accepted communications.

2. The system of claim 1 further comprising:

means for removing from said delivered ones of said communications said additional information not part of said protocol.

3. The method of claim 1 wherein said additional information is contained as part of metadata in said protocol.

4. The system of claim 1 wherein said additional information includes the identity of a particular transaction, said system further comprising:

means for using said stored data to resolve error situations within a particular transaction.

5. The system of claim 4 further comprising:

means for tuning said IVR based upon said stored communications, including both said control protocol and said voice stream.

6. The system of claim 1 further comprising:

means for changing the rules base of said proxy from time to time.

7. The system of claim 6 wherein one of said rules instructs said proxy to direct certain of said communications for alternate resolution, said alternate resolution selected from the list of autotuning, transcription, system management, metrics processing, media processing, selection of speech servers.

8. An IVR system comprising:

an application environment comprising;

at least one speech server for providing voice prompts to, and interpreting voice responses from, callers via a communication network;

a browser for interfacing said communication network with said speech server, said browser operating from instructions provided from said application environment, said instructions including a session identification for attachment to all communications between said browser and said speech server;

said communication including both control and media data, said application environment further comprising;

a proxy for intercepting communications between said browser and said media server, said intercepted communications being stored in said database;

at least one database for storing said removed information and said intercepted communications in said database in association with a common session identification; and

at least one application for accepting from said database a stored record of a particular session and for performing a transformation of said session stored media data.

9. The IVR system of claim 8 wherein said proxy is further operable for removing from said communications information not part of a protocol used for control between said browser and said media server.

10. The IVR system of claim 8 wherein said session identification is contained in metadata in communications between said browser and said media server.

11. The IVR system of claim 8 wherein said proxy further comprises:

a process for intercepting certain communications from said media server and for substituting a modified communication for said intercepted communication.

12. The IVR system of claim 11 wherein said intercepted communication is an error message from said media server and wherein said modified communication is a correct response to a media server request.

13. The IVR of claim 11 wherein said correct response is generated during a single transaction between said browser and said media server.

14. The IVR system of claim 8 wherein said transformation is selected from the list of:

speech to text transcription;

storage on a storage medium individual to said particular transaction;

interpretation of graphical images.

15. A method for collecting data in a voice response system, said method comprising:

adding session identification to each communication to and from a speech server;

capturing communications to and from said speech server;

each captured communication including said removed added session identification; and

making any said captured communications available for additional processing on a session by session basis.

16. The method of claim 15 further comprising:

removing said added session identification on each said communication.

17. The method of claim 15 wherein said session identification is added as part of metadata on each said communication.

18. The method of claim 17 wherein said communication to and from said speech server is from a browser using the MRCP communication protocol and wherein said added session data is hidden within said protocol.

19. The method of claim 17 wherein said additional processing comprises:

performing system functions using said captured communications, said system functions selected from the list of;

autotuning, transcription services, metric analysis, management functions, selection of speech servers, graphical interpretations, recording onto a portable medium.

20. The method of claim 19 further comprising:

incorporating metadata with said captured communications, said metadata used to enhance the performance of said system functions.

21. The method of claim 17 further comprising:

transcribing portions of said captured communications.

22. The method of claim 17 further comprising:

retrieving all said captured communications to and from said speech server for a selected set of session IDs;

translating said captured communications into a human recognizable format; and

recording said translated format in a storage media.

23. The method of claim 22 wherein said storage media is portable.

24. An IVR system comprising:

a browser;

a speech server;

a database; and

means for collecting attribute data from messages passing on a communication path between said browser and said server, said attribute data pertaining to commands, events and command results.

25. The IVR system of claim 24 further comprising:

means for stripping from said messages passing on said communication path any data added to said messages for session identification purposes.

26. The IVR system of claim 24 wherein any said added data is hidden from said speech server within and foreign to a protocol used for such communication path communications.

27. The method of logging data in an IVR system, said method comprising:

sending messages between a browser and a speech server, said messages pertaining to commands, events and data from said browser, from said speech server and from application programs running on either;

incorporating in said messages information relating to the session in which said commands, events and data belong;

extracting from sent ones of said messages said incorporated information;

storing said extracted messages together with said extracted session information; and

translating at least a portion of said stored extracted data on a session by session basis.

28. The method of claim 27 further comprising:

selecting one of a plurality of possible speech servers for each said sent message, said selecting based, at least in part, on said stored extracted messages.

29. The method of claim 27 further comprising:

invoking the assistance of a process other than said speech server process, said invoking being triggered by data contained in a message between said speech server and said browser, said invoking including translating data associated with a particular message.

30. The method of claim 27 further comprising:

invoking the assistance of a process other than said speech server process, said invoking being triggered by data contained in a group of messages between said speech server and said browser, said message group sharing a common session ID as contained in said stored extracted messages;

said invoking including translating data associated with a particular message.