System and method for parameterization of speech recognition grammar specification (SRGS) grammars

1. A method for automatically recognizing speech matching grammars in audio, comprising:

loading, by a processor, a grammar specification defining at least one parameterizable grammar, the at least one parameterizable grammar corresponding to a type of speech input and the at least one parameterizable grammar comprising a plurality of rules defining constraints on speech corresponding to the type of speech input;

setting, by the processor, an initial state of a grammar processor as a current state, the current state comprising parameters supplied to the rules;

selecting, by the processor, a rule of the plurality of rules matching the parameters of the current state of the grammar processor;

applying, by the processor, the selected rule to the audio and updating the current state to compute an updated state comprising a plurality of updated parameters;

determining, by the processor, whether termination conditions have been met;

in response to determining the termination conditions are not met, selecting, by the processor, from the plurality of rules in accordance with the updated parameters of the updated state; and

in response to determining the termination conditions are met, outputting, by the processor, a recognizer result of the updated state.

2. The method of claim 1 , wherein each of the rules comprises a corresponding matching condition, the matching conditions having different levels of narrowness, the rules being ordered in accordance with corresponding levels of narrowness.

3. The method of claim 2 , wherein the selecting the rule from the plurality of rules comprises:

selecting a narrowest rule of the rules as a current rule; and

iteratively:

determining if the matching condition of the current rule is met;

in response to determining that the matching condition is met, applying the current rule; and

in response to determining that matching condition is not met, setting a next rule of the rules, in order of narrowness, as the current rule.

4. The method of claim 3 , wherein the matching condition comprises a Boolean expression.

5. The method of claim 4 , wherein the Boolean expression comprises an inequality condition.

6. The method of claim 1 , wherein the grammar is a dual-tone multi-frequency grammar.

7. The method of claim 1 , wherein the grammar is a voice grammar.

8. A system comprising:

a processor; and

memory having instructions stored thereon that, when executed by the processor, cause the processor to automatically recognize speech matching grammars in audio by:

loading a grammar specification defining at least one parameterizable grammar, the at least one parameterizable grammar corresponding to a type of speech input and the at least one parameterizable grammar comprising a plurality of rules defining constraints on speech corresponding to the type of speech input;

setting an initial state of a grammar processor as a current state, the current state comprising parameters supplied to the rules;

selecting a rule of the plurality of rules matching the parameters of the current state of the grammar processor;

applying the selected rule to the audio and updating the current state to compute an updated state comprising a plurality of updated parameters;

determining whether termination conditions have been met;

in response to determining the termination conditions are not met, selecting from the plurality of rules in accordance with the updated parameters of the updated state; and

in response to determining the termination conditions are met, outputting a recognizer result of the updated state.

9. The system of claim 8 , wherein each of the rules comprises a corresponding matching condition, the matching conditions having different levels of narrowness, the rules being ordered in accordance with corresponding levels of narrowness.

10. The system of claim 9 , wherein the memory further stores instructions that, when executed by the processor, cause the processor to select a rule from the plurality of rules by:

selecting a narrowest rule of the rules as a current rule; and

iteratively:

determining if the matching condition of the current rule is met;

in response to determining that the matching condition is met, applying the current rule; and

in response to determining that matching condition is not met, setting a next rule of the rules, in order of narrowness, as the current rule.

11. The system of claim 10 , wherein the matching condition comprises a Boolean expression.

12. The system of claim 11 , wherein the Boolean expression comprises an inequality condition.

13. The system of claim 8 , wherein the grammar is a dual-tone multi-frequency grammar.

14. The system of claim 8 , wherein the grammar is a voice grammar.

15. A system for automatically recognizing speech matching grammars in audio, comprising:

means for loading a grammar specification defining at least one parameterizable grammar, the at least one parameterizable grammar corresponding to a type of speech input and the at least one parameterizable grammar comprising a plurality of rules defining constraints on speech corresponding to the type of speech input;

means for setting an initial state of a grammar processor as a current state, the current state comprising parameters supplied to the rules;

means for selecting a rule of the plurality of rules matching the parameters of the current state of the grammar processor;

means for applying the selected rule to the audio and updating the current state to compute an updated state comprising a plurality of updated parameters;

means for determining whether termination conditions have been met;

means for selecting from the plurality of rules in accordance with the updated parameters of the updated state, in response to determining the termination conditions are not met; and

means for outputting a recognizer result of the updated state in response to determining the termination conditions are met.

16. The system of claim 15 , wherein each of the rules comprises a corresponding matching condition, the matching conditions having different levels of narrowness, the rules being ordered in accordance with corresponding levels of narrowness.

17. The system of claim 16 , wherein the means for selecting a rule from the plurality of rules comprises:

means for selecting a narrowest rule of the rules as a current rule; and

means for, iteratively:

determining if the matching condition of the current rule is met;

in response to determining that the matching condition is met, applying the current rule; and

in response to determining that matching condition is not met, setting a next rule of the rules, in order of narrowness, as the current rule.

18. The system of claim 17 , wherein the matching condition comprises a Boolean expression.

19. The system of claim 18 wherein the Boolean expression comprises an inequality condition.