Method and system for enhanced response to voice commands in a voice command platform

• US 6,985,865 B1
• Filed: 09/26/2001
• Issued: 01/10/2006
• Est. Priority Date: 09/26/2001
• Status: Expired due to Term

• Current Assignee: Sprint Spectrum LP (Deutsche Telekom AG)
• Original Assignee: Sprint Spectrum LP (Deutsche Telekom AG)
• Inventors: Roche, Elizabeth, Packingham, Kevin
• Primary Examiner(s): Young, W. R.
• Assistant Examiner(s): Sked, Matthew J
• Application Number:
  US09/964,117
• Time in Patent Office: 1,567 Days
• Field of Search: 704/257, 704/251, 704/270.1, 704/275
• US Class Current: 704/275
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command
• CPC Class Codes: G10L 2015/228 of application context

Abstract

A method and system for responding to voice commands in a voice-command platform. The platform receives a voice command during a voice command session with a user, and the platform responsively carries out a set of system-level logic. For example, the platform may determine that the voice command is an alias-grammar (whether personal to a user, or global for the platform) and may then convert the alias-grammar into an actual grammar. As another example, the platform may determine that the voice command is a bookmark command, such as a request to save a given voice command navigation point in a centralized list of bookmarks for the user, or to recall a navigation point from the user'"'"'s centralized list, and the platform may respond to the bookmark command accordingly. And as yet another example, the platform may recognize a voice command that is not defined by a voice command application currently being processed but that is instead defined globally for the voice command platform.

Claims

1. A voice command system comprising:

a user communication interface for communicating with users via a telecommunications network;

a processor;

an application-processing module executable by the processor to process voice command applications, the voice command applications defining allowed grammars and application logic;

a voice-processing module executable by the processor to recognize grammars in speech signals received from a user via the user communication interface; and

aliasing-logic executable by the processor, upon recognition of an alias grammar in a speech signal received from the user, to convert the alias grammar to an actual grammar, and to recognize the actual grammar as an allowed grammar defined by a voice command application;

wherein the aliasing-logic uses a predefined set of correlations between alias grammars and actual grammars to convert between an alias grammar and an actual grammar; and

wherein the processor uses at least a subset of the predefined set of correlations regardless of the user who is communicating with the voice command system and substantially regardless of which voice command application the processor is processing.

2. The voice command system of claim 1, wherein the aliasing-logic comprises user profile data that indicates, respectively for each of a plurality of users, correlations between alias grammars and actual grammars.

3. The voice command system of claim 2, wherein:

the user profile data correlates a given actual grammar with a first alias grammar for a first user; and

the user profile data correlates the given actual grammar with a second alias grammar for a second user.

4. The voice command system of claim 2, further comprising:

a user profile store containing the user profile data,

personalization-logic executable by the processor for retrieving from the user profile store the correlations between alias grammars and actual grammars for the user.

5. The voice command system of claim 4, wherein the personalization-logic is executable by the processor to retrieve the correlations during a voice command session with the user.

6. The voice command system of claim 5, wherein the personalization-logic is executable by the processor to retrieve the correlations at the initiation of the voice command session with the user.

7. The voice command system of claim 1, further comprising a memory, wherein the user profile data for a given user is stored in memory in the platform during a voice command session with the given user.

8. The voice command system of claim 1, further comprising provisioning-logic for receiving a set of user-defined correlations between alias grammars and actual grammars.

9. The voice command system of claim 8, wherein the provisioning-logic comprises a web interface accessible by a user via a computer network.

10. The voice command system of claim 1, wherein the telecommunications network comprises a wireless communications link.

11. A voice command system comprising:

a user communication interface for communicating with users via a telecommunications network;

a processor;

an application-processing module executable by the processor to process voice command applications, the voice command applications defining allowed grammars and application logic;

a voice-processing module executable by the processor to recognize grammars in speech signals received from a user via the user communication interface; and

aliasing-logic executable by the processor, upon recognition of an alias grammar in a speech signal received from the user, to convert the alias grammar to an actual grammar, and to recognize the actual grammar as an allowed grammar defined by a voice command application, wherein the aliasing-logic comprises user profile data that indicates, respectively for each of a plurality of users, correlations between alias grammars and actual grammars; and

wherein the processor executes the aliasing logic substantially regardless of which voice command application the processor is currently processing.