Speech processing for telephony API

• US 7,634,066 B2
• Filed: 07/01/2004
• Issued: 12/15/2009
• Est. Priority Date: 09/21/1998
• Status: Expired due to Term

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Microsoft Corporation
• Inventors: Quinton, Mary Michelle
• Primary Examiner(s): Tsang; Fan
• Assistant Examiner(s): Islam; Mohammad K
• Application Number:
  US10/882,703
• Publication Number:
  US 20040240629A1
• Time in Patent Office: 1,993 Days
• Field of Search: 379/67.1, 379 8813- 8818, 379/355.07, 704/270
• US Class Current: 379/88.17
• CPC Class Codes: H04L 65/1101 Session protocols
• CPC Class Codes: H04L 65/401 wherein the services involve a main real-time session and one or more additional parallel real-time or time sensitive sessions, e.g. white board sharing or spawning of a subconference
• CPC Class Codes: H04M 1/2473 Telephone terminals interfacing a personal computer, e.g. using an API (Application Programming Interface) details of API H04M7/0021
• CPC Class Codes: H04M 1/2478 Telephone terminals specially adapted for non-voice services, e.g. email, internet access centralized arrangements where telephones services are combined H04M7/0024; for recording text messages H04M3/5322; for accessing Internet H04M3/4938
• CPC Class Codes: H04M 3/42323 PBX's with CTI arrangements

Abstract

Systems, methods, and structures are discussed that enhance media processing. One aspect of the present invention includes a data structure to enhance media processing. The data structure includes a terminal data structure to instantiate terminal objects and a speech recognition terminal data structure that extends the terminal data structure. Another aspect of the present invention includes a data structure to enhance media processing. This data structure includes a terminal data structure to instantiate terminal objects and a speech generation terminal data structure that extends the terminal data structure. These data structures may be used to implement an internet protocol interactive voice response system, an internet protocol unified message system, and speech-enabled Web applications.

Claims

1. A computing device having a memory and a processor for enhancing media processing of a media stream containing speech data, comprising:

a terminal data structure to support instantiating terminal objects, each terminal object adhering to a uniform interface, providing a telephony service, and having a terminal class name and a media type;

a speech recognition terminal data structure that extends the terminal data structure;

a terminal manager to instantiate, based on the terminal data structure and the speech recognition terminal data structure, terminal objects including a speech recognition terminal object to recognize speech having a speech recognition media type; and

a TAPI application component providing a telephony API to form a connection with a client and to process the speech data by;

registering terminal objects including a speech recognition terminal object;

selecting, with the processor, the speech recognition terminal object from among a group of registered terminal objects based on the media type of the registered terminal objects;

instantiating the selected speech recognition terminal object using the terminal manager by providing the terminal class name, the media type, and a method of signaling events; and

providing the speech data to the instantiated speech recognition terminal object for recognition of the speech data.

2. The computing device of claim 1, wherein the speech recognition terminal data structure includes an engine token data structure.

3. The computing device of claim 1, wherein the speech recognition terminal data structure includes an enumeration engine data structure.

4. The computing device of claim 1, wherein the speech recognition terminal data structure includes a speech recognition data structure.

5. The computing device of claim 1, wherein the speech recognition terminal data structure includes a recognition context data structure.

6. A computer-readable storage medium containing computer-executable instructions that when executed by a computer having a memory and a processor cause the computer to perform a method for enhancing media processing, the method comprising:

providing a terminal data structure to support instantiating terminal objects, each terminal object adhering to a uniform interface, providing a telephony service, and having a terminal class name and a media type;

providing a speech recognition terminal data structure that extends the terminal data structure, wherein the speech recognition terminal data structure includes an engine token data structure; and

using a telephony API provided by a TAPI application component to form a connection with a client and to process speech data by;

registering terminal objects including a speech recognition terminal object;

selecting, with the processor, the speech recognition terminal object from among a group of registered terminal objects based on the media type of the registered terminal objects;

instantiating the selected speech recognition terminal object using a terminal manager, based on the terminal data structure and the speech recognition terminal data structure, by providing the terminal class name, the media type, and a method of signaling events; and

providing speech data to the instantiated speech recognition terminal object for recognition of the speech data.

7. The computer-readable storage medium of claim 6, wherein the engine token data structure includes a method member get engine name for getting a name of a speech recognition engine in a textual form.

8. The computer-readable storage medium of claim 6, wherein the engine token data structure includes a method member get engine token for getting an identifier that identifies a speech recognition engine.

9. A computing system having a memory and a processor for enhancing media processing of speech, comprising:

a terminal data structure to support instantiating terminal objects, each terminal object adhering to a uniform interface, providing a telephony service, and having a terminal class name and a media type;

a speech recognition terminal data structure that extends the terminal data structure, wherein the speech recognition terminal data structure includes an enumeration engine data structure; and

a TAPI application component providing a telephony API to form a connection with a client in response to receiving a call from the client and to process the speech by;

registering terminal objects including a speech recognition terminal object;

selecting, with the processor, the speech recognition terminal object from among a group of registered terminal objects based on media type of the registered terminal objects;

instantiating the selected speech recognition terminal object using a terminal manager, bsed on the terminal data structure and the speech recognition terminal data structure, by providing the terminal class name, the media type, and a method of signaling events; and

providing the speech to the instantiated speech recognition terminal object for recognition of the speech data.

10. The computing system of claim 9, wherein the enumeration engine data structure includes a method member next for getting a next available speech recognition engine.

11. A computing system having a memory and a processor for enhancing media processing, comprising:

a terminal data structure to support instantiating terminal objects, each terminal object adhering to a uniform interface, providing a telephony service, and having a terminal class name and a media type;

a speech recognition terminal data structure that extends the terminal data structure, wherein the speech recognition terminal data structure includes a speech recognition data structure; and

a TAPI application component providing a telephony API to form a connection with a client in response to receiving a call from the client, and to process speech by;

registering terminal objects including a speech recognition terminal object;

selecting, with the processor, the speech recognition terminal object from among a group of registered terminal objects;

instantiating the selected speech recognition terminal object using a terminal manager, based on terminal data structure and the speech recognition terminal data structure, by providing the terminal class name, the media type, and a method of signaling events; and

providing speech to the instantiated speech recognition terminal object for recognition of the speech.

12. The computing system of claim 11, wherein the speech recognition data structure includes a member method enumerate recognition engines for obtaining an indirect reference to a listing of speech recognition engines that are available for use.

13. The computing system of claim 11, wherein the speech recognition data structure includes a member method select engine for selecting a speech recognition engine to be used.

14. The computing system of claim 11, wherein the speech recognition data structure includes a member method get selected engine for retrieving the currently selected speech recognition engine.

15. The computing system of claim 11, wherein the speech recognition data structure includes a member method convert extended markup language to grammar for converting extended markup language text into a compiled grammar for use with a speech recognition engine.

16. A computer-readable storage medium containing computer-executable instructions that when executed by a computer having a memory and a processor cause the computer to perform a method for enhancing media processing of speech, the method comprising:

receiving a terminal data structure to support instantiating terminal objects, each terminal object adhering to a uniform interface, providing a telephony service, and having a terminal class name and a media type;

receiving a speech recognition terminal data structure that extends the terminal data structure, wherein the speech recognition terminal data structure includes a recognition context data structure; and

providing by a TAPI application component a telephony API to form a connection with a client and to process the speech by;

registering terminal objects including a speech recognition terminal object;

selecting, with the processor, the speech recognition terminal object from among a group of registered terminal objects based on the media type of the registered terminal objects;

instantiating the selected speech recognition terminal object using a terminal manager, based on the terminal data structure and speech recognition terminal data structure, by providing the terminal class name, the media type, and a method of signaling events; and

providing speech data to the instantiated speech recognition terminal object for recognition of the speech data.

17. The computer-readable storage medium of claim 16, wherein the recognition context data structure includes a method member initialize for creating a speech recognition context based on a selected speech recognition engine.

18. The computer-readable storage medium of claim 16, wherein the recognition context data structure includes a method member shut down for destroying a speech recognition context.

19. The computer-readable storage medium of claim 16, wherein the recognition context data structure includes a method member load grammar for loading a grammar into a recognition context from a source selected from a group consisting of a resource, a memory, and a file.

20. The computer-readable storage medium of claim 16, wherein the recognition context data structure includes a method member unload grammar for unloading a grammar previously loaded into a recognition context.

21. The computer-readable storage medium of claim 16, wherein the recognition context data structure includes a method member activate grammar for activating a grammar to be used in a speech recognition engine.

22. The computer-readable storage medium of claim 16, wherein the recognition context data structure includes a method member get result for retrieving a speech recognition result.

23. The computer-readable storage medium of claim 16, wherein the recognition context data structure includes a method member get hypothesis for retrieving a speech recognition result that is deemed a likely speech recognition result.

24. A method in a computing device having a memory and a processor for enhancing media processing, comprising:

providing terminal objects adhering to a uniform interface, providing telephony services, and having terminal class names and media types, the terminal objects including a speech recognition terminal object for recognizing speech;

with a processor, invoking a telephony API to listen for incoming calls, wherein the telephony API is provided by a TAPI application component, the TAPI application component for;

registering terminal objects that are selectable for instantiation with a terminal manager by providing a terminal class name, a media type, and a method of signaling events and

providing a list of registered terminal objects including the media types of the registered terminal objects;

upon receiving a call from a client, with the processor, invoking the telephony API to form a connection with the client and to select the speech recognition terminal object from among a group of registered terminal objects;

with a processor, invoking a method of the selected speech recognition terminal object to get a desired speech recognition engine; and

setting a speech recognition context based on the desired speech recognition terminal object to recognize speech associated with the received call.

25. The method of claim 24, further comprising:

enumerating a list of available speech recognition engines.

26. The method of claim 25, further comprising:

identifying a desired speech recognition engine from the list of available speech recognition engines.

27. The method of claim 26, further comprising:

selecting the desired speech recognition engine.

28. The method of claim 24, wherein setting includes initializing the speech recognition context.

29. The method of claim 24, wherein setting includes loading a grammar for the speech recognition context.

30. The method of claim 24, wherein setting includes activating a grammar for the speech recognition context.

31. The method of claim 24, wherein setting includes setting the speech recognition context to notify a user when a desired event occurs.

33. The method of claim 24, further comprising:

invoking a method of the speech recognition context to retrieve a speech recognition result.

32. A system for enhancing processing of speech of a media stream, comprising:

a processor;

a speech terminal object to recognize the speech of the media stream; and

a TAPI application component providing a telephony API to form a connection with a client and to process the speech of the media stream by;

registering terminal objects including the speech terminal object, each terminal object adhering to a uniform interface;

selecting the speech terminal object from among a group of registered terminal objects;

instantiating the selected speech terminal object using the terminal manager by providing the terminal class name, the media type, and a method of signaling events; and

providing speech of the media stream to the instantiated speech terminal object for recognition of the speech of the media stream

wherein the speech terminal object includes computer-executable instructions stored in a memory for execution by the processor.