Method and system of VoiceXML interpreting

• US 7,174,006 B2
• Filed: 06/18/2001
• Issued: 02/06/2007
• Est. Priority Date: 06/18/2001
• Status: Active Grant

• Current Assignee: Dialogic, Inc. (Enghouse Systems Limited)
• Original Assignee: NMS Communications Corporation (Live Microsystems, Inc.)
• Inventors: Guedalia, David, Reisler, Lawrence, Raanan, Gavriel
• Primary Examiner(s): Gauthier, Gerald
• Application Number:
  US09/883,757
• Publication Number:
  US 20020191756A1
• Time in Patent Office: 2,059 Days
• Field of Search: 379/88.13, 379/88.04, 379/93.23, 379/88.01, 707/513, 717100-175, 704/270, 704/275, 712/209
• US Class Current: 379/88.13
• CPC Class Codes: H04M 3/4938 comprising a voice browser which renders and interprets, e.g. VoiceXML
• CPC Class Codes: Y10S 707/99953 Recoverability
• CPC Class Codes: Y10S 707/99957 Garbage collection

Abstract

A VoiceXML interpreting system and method including a VoiceXML Interpreter. The VoiceXML interpreter includes a Fetcher operative to retrieve documents, a compiler operative to compile documents retrieved by the Fetcher and a cache which stores compiled documents compiled by the compiler.

Claims

1. An interactive voice response system comprising:

a compiler operative to compile documents retrieved by a fetcher into compiled document data in executable form;

a cache which stores the compiled document data prior to receipt of audio input from a given user requesting a text-based document;

an execution thread that executes compiled document data retrieved from the cache by the fetcher;

a storage device, which stores state information related to execution of said compiled document data; and

a backup VoiceXML Interpreter communicating with said storage device, the backup interpreter providing a response to a user in the event of a failure associated with a primary voice response system.

2. A system according to claim 1, wherein said storage device comprises a memory database external to said backup VoiceXML Interpreter.

3. In an interactive voice response system, a method comprising:

retrieving documents encoded according to VoiceXML;

compiling the retrieved documents into compiled document data in executable form;

caching the compiled document data for later retrieval and execution;

storing state information related to execution of said compiled document data; and

providing a backup VoiceXML Interpreter that utilizes the stored state information to support continued service in the event of a failure.

4. An interactive voice response system configured as a server that provides requested audio information associated with text-based documents, the server comprising:

an execution thread that processes an incoming request and, based on an audio input from a given user, identifies a request for audio information associated with a text-based document; and

a fetcher that receives a signal from the execution thread to search a cache for executable code associated with the requested audio information, the fetcher retrieving corresponding executable code from the cache for execution by the execution thread to satisfy the request for audio information associated with the incoming request;

wherein the fetcher initiates communication with a remote server to retrieve a text-based document associated with the requested information if corresponding executable code is not stored in the cache;

a compiler that converts retrieved text-based documents into executable code for storage in the cache; and

the server coupled to a database that stores state information associated with executable code being executed by the execution thread, the state information accessible by a backup interpreter to provide interactive voice response service in the event of a failure associated with the interactive voice response system.

5. A server as in claim 4 further comprising:

a compiler that converts the text-based document into executable speech code for storage in the cache prior to receipt of the incoming request.

6. A server as in claim 4, wherein executable code stored in the cache is concurrently utilized by multiple execution threads to provide a response to multiple users.

7. A server as in claim 4 further comprising:

a storage device to store state information related to the executable code executed by the execution thread to satisfy the request for audio information associated with the incoming request.

8. A server as in claim 4, wherein the executable code retrieved from the cache is associated with a corresponding viewable text-based document available on the World Wide Web.

9. A server as in claim 4, wherein the execution thread receives the incoming call from a switchboard.

10. A server as in claim 4 configured as a VoiceXML interpreter.

11. A server as in claim 5, wherein the compiler converts the text-based document into executable code in response to a previous request by another user for the text-based document.

12. A server as in claim 4, wherein the executable code stored in the cache is used at a later time to respond to corresponding future requests with respect to the text-based documents.

13. A server as in claim 4, wherein the backup interpreter is a backup VoiceXML interpreter.

14. A method of providing requested audio information associated with text-based documents, the method comprising:

processing an incoming call based on an audio input from a given user;

identifying a request for audio information associated with a text-based document;

searching a cache for executable code associated with the requested audio information, the executable code generated in response to a previous request from another user for audio information associated with the text-based document;

executing corresponding executable code from the cache to satisfy the request for audio information associated with the incoming call; and

implementing a backup interpreter to process a subsequent incoming call if a primary interactive voice response system happens to fail during processing of the subsequent incoming call.

15. A method as in claim 14, wherein searching the cache for executable code associated with the requested audio information includes:

searching amongst multiple sets of stored executable code in the cache for the executable code associated with the requested information, the multiple sets of executable code in the cache corresponding to documents previously compiled as a result of other respective users previously requesting audio information associated with the documents.

16. A method as in claim 14, wherein implementing the backup interpreter includes implementing a VoiceXML backup interpreter to process the subsequent incoming call.

17. An interactive voice response system that provides requested audio information associated with text-based documents, the interactive voice response system comprising:

an execution thread that processes an incoming request and, based on an audio input from a given user, identifies a request for audio information associated with a text-based document;

a cache that stores sets of executable code associated with corresponding previously compiled documents that can be requested from the given user;

a fetcher that receives a signal from the execution thread to search the cache for executable code associated with the requested audio information, the fetcher conditionally retrieving corresponding executable code from the cache for execution by the execution thread to satisfy the request for audio information associated with the incoming and

the interactive voice response system coupled to a database that stores state information associated with executable code being executed by the execution thread, the state information accessible by a backup interpreter to provide interactive voice response service during a failure associated with the interactive voice response system.

18. An interactive voice response system as in claim 17, wherein the corresponding executable code is stored in the cache as a result of a previous request with respect to the text-based document, the fetcher searching the cache in order to retrieve the corresponding executable code from the cache for execution of the corresponding execution code by the execution thread to satisfy the request for audio information associated with the incoming request.

19. An interactive voice response system as in claim 18, wherein the corresponding executable code is executable speech code, the interactive voice response system further comprising:

a compiler that converts the text-based document into the executable speech code for storage in the cache prior to receipt of the incoming request.

20. An interactive voice response system as in claim 19, wherein the fetcher initiates communication with a remote server to retrieve a text-based document associated with the requested information over a network connection if the corresponding executable code were not stored in the cache.

21. An interactive voice response system as in claim 19, wherein the corresponding executable code fetched from the cache is concurrently utilized by multiple execution threads to provide a response to multiple users for requests with respect to a same text-based document.

22. An interactive voice response system as in claim 19, wherein the corresponding executable code retrieved from the cache is associated with a corresponding viewable text-based document accessible over the World Wide Web.

23. An interactive voice response system as in claim 17, wherein the backup interpreter processes incoming requests by interpreting retrieved documents that are coded according to an interactive voice application programming language.