US 2008/0288259 A1 · SPEECH RECOGNITION MACRO RUNTIME

Abstract

The disclosed speech recognition system enables users to define personalized, context-aware voice commands without extensive software development. Command sets may be defined in a user-friendly language and stored in an eXtensible Markup Language (XML) file. Each command object within the command set may include one or more user configurable actions, one or more configurable rules, and one or more configurable conditions The command sets may be managed by a command set loader, that loads and processes each command set into computer executable code. The command set loader may enable and disable command sets. A macro processing component may provide a speech recognition grammar to an API of the speech recognition engine based on currently enabled commands. When the speech recognition engine recognizes user speech consistent with the grammar, the macro processing component may initiate the one or more computer executable actions.

Claims20

Dependent claims

A method of using a speech recognition engine, the method comprising:

receiving a first command object, wherein the first command object comprises first data indicative of a word, a condition, and a first computer executable action;
evaluating the condition;
generating a first speech recognition grammar comprising the word in accordance with an outcome of the evaluating;
communicating the first speech recognition grammar to the speech recognition engine;
and
executing the first computer executable action responsive to receiving a recognition indication from the speech recognition engine, wherein the recognition indication is associated with the word.

3 more elements

Dependencies are inferred from the claim text.

Speech recognition macro runtime

Life of this patent

Speech recognition macro runtime

Life of this patent

Abstract

Claims20