System and method for providing remote automatic speech recognition and text to speech services via a packet network

1. A method of operating an automatic speech recognition application remotely accessible over an Internet Protocol network, the method comprising:

receiving over an Internet Protocol network an identification of an automatic speech recognition application;

selecting a grammar from a plurality of grammars according to the identified automatic speech recognition application;

receiving speech information over the Internet Protocol network; and

using the selected grammar, recognizing the received speech information.

2. The method of claim 1 , further comprising sending information based on the recognized speech over the Internet Protocol network to a client.

3. The method of claim 2 , wherein the plurality of grammars is stored local to the automatic speech recognition service.

4. The method of claim 2 , wherein the plurality of grammars is stored remotely from the automatic speech recognition service.

5. The method of claim 1 , wherein the identified automatic speech recognition application is identified by a universal resource locator (URL) compatible address of the automatic speech recognition application.

6. The method of claim 1 , wherein the automatic speech recognition application is a component of a spoken dialog service.

7. The method of claim 6 , wherein the spoken dialog application transmits synthetic speech to a client over the Internet Protocol network to carry on a dialog with a user communicating with the client.

8. The method of claim 1 , further comprising:

identifying a sub-portion of a dialog with a user;

selecting a sub-grammar according to the identified sub-portion of the dialog; and

recognizing speech from the user for the identified sub-portion of the dialog using the selected sub-grammar.

9. The method of claim 8 , wherein the sub-portion of the dialog relates to a task.

10. The method of claim 1 , wherein the selected grammar comprises a plurality of sub-grammars and each sub-grammar is selected and used to recognize speech from a user according to a portion of a dialog with a user.

11. The method of claim 10 , wherein each sub-grammar is associated with a task.

12. A method of operating a spoken dialog service on a server remotely accessible over an Internet Protocol network, the method comprising, at the server:

receiving over the Internet Protocol network an identification of a spoken dialog application having an associated grammar;

receiving speech information from a user;

transmitting the speech information to the server;

recognizing the speech using the associated grammar; and

transmitting generated speech from the server in response to the recognized user speech information over the Internet Protocol network.

13. The method of claim 12 , further comprising:

identifying a sub-portion of a dialog with a user;

selecting a sub-grammar according to the identified sub-portion of the dialog; and

recognizing speech from the user for the identified sub-portion of the dialog using the selected sub-grammar.

14. The method of claim 13 , wherein each sub-grammar is associated with a task.

15. The method of claim 12 , wherein the associated grammar is identified by a universal resource locator (URL) compatible address associated with the spoken dialog application.

16. A system for operating an automatic speech recognition application remotely accessible over an Internet Protocol network, the system comprising:

means for receiving over an Internet Protocol network an identification of an automatic speech recognition application;

means for selecting a grammar from a plurality of grammars according to the identified automatic speech recognition application;

means for receiving speech information over the Internet Protocol network; and

means for using the selected grammar to recognize the received speech information.

17. A system for operating an automatic speech recognition application remotely accessible over an Internet Protocol network, the system comprising:

a module configured to receive over an Internet Protocol network an identification of an automatic speech recognition application;

a module configured to select a grammar from a plurality of grammars according to the identified automatic speech recognition application;

a module configured to receive speech information over the Internet Protocol network; and

a module that uses the selected grammar to recognize the received speech information.

18. The system of claim 17 , wherein the system comprises a spoken dialog service comprising the automatic speech recognition application.

19. A computer-readable medium storing instructions for controlling a computing device to operate an automatic speech recognition application remotely accessible over an Internet Protocol network, the instructions comprising:

receiving over an Internet Protocol network an identification of an automatic speech recognition application;

selecting a grammar from a plurality of grammars according to the identified automatic speech recognition application;

receiving speech information over the Internet Protocol network; and

using the selected grammar, recognizing the received speech information.

20. The computer-readable medium of claim 19 , wherein the automatic speech recognition application is part of a spoken dialog service.