Method and apparatus for voice recognition
View Patent ↗A voice recognition system according to embodiments of the present invention includes a recognizer; a grammar operably coupled to the recognizer, the grammar defining utterances to be recognized; a processing application configured to receive an output of the recognizer, the output including a tag element having a semantic content and including an embedded instruction for said processing application, such that said processing application can process the instruction independent of a semantic content of the tag element.
1. A voice recognition system, comprising:
a recognizer configured to receive a voice input;
a grammar operably coupled to the recognizer, the grammar defining utterances to be recognized, the recognizer encoding tags for said received voice input responsive to defined said utterances;
a processing application configured to receive encoded said tags, each encoded tag having a semantic content and including an embedded function for said processing application, such that said processing application can process the function independent of a semantic content of the tag, each encoded tag including a grammar identification (ID), a tag type and data.
2. A voice recognition system in accordance with claim 1 , wherein said processing application is a VXML application, the grammar ID identifies a particular grammar and tag types include link, embedded prompt, and non-embedded prompt.
3. A voice recognition system in accordance with claim 2 , wherein the tag type is embedded prompt and said data identifies an audio file to be played.
4. A voice recognition system in accordance with claim 2 , wherein the tag type is link and said data identifies an URL.
5. A voice recognition system in accordance with claim 2 , wherein the tag type is non-embedded prompt and said data provides a prompt.
6. A voice recognition system in accordance with claim 2 , wherein the grammar is an XP grammar recognizing utterances that include multiple pieces of information.
7. A method, comprising:
providing a recognizer adapted to receive voice input;
providing a grammar programming unit, the grammar programming unit adapted for encoding instructions and semantic content in a tag in cooperation with the recognizer, tags encoded for the received voice input responsive to defined utterances, each encoded tag including a grammar identification (ID), a tag type and data;
providing the encoded tags to a processing application configured to read and execute said instructions without processing said semantic content of said encoded tag.
8. A method in accordance with claim 7 , wherein said processing application includes a VXML browser, the grammar ID identifies a particular grammar and tag types include link, embedded prompt, and non-embedded prompt.
9. A method in accordance with claim 7 , wherein providing a processing application further comprises providing a processing application configured to be interchangeably usable with a plurality of grammars, each identified by a particular said grammar ID.
10. A method in accordance with claim 9 , wherein the tag type is embedded prompt and said data identifies an audio file.
11. A method in accordance with claim 9 , wherein tag type is non-embedded prompt and said data identifies a prompt.
12. A method in accordance with claim 9 , wherein the tag type is link and said data is an URL.
13. A method in accordance with claim 9 , wherein the plurality of grammars include XP grammars recognizing utterances that include multiple pieces of information.
14. A telecommunications system, comprising:
a network;
a plurality of user devices operably coupled to the network;
a voice recognition unit accessible by said user devices, said voice recognition unit including a recognizer configured to receive a voice input;
a grammar operably coupled to the recognizer, the grammar defining utterances to be recognized, the recognizer encoding tags for said received voice input responsive to defined said utterances;
a processing application configured to receive encoded said tags, each encoded tag having a semantic content and including at least one embedded function for said processing application, such that said processing application can process the function independent of a semantic content of the tag, each encoded tag including a grammar identification (ID), a tag type and data.
15. A telecommunications system in accordance with claim 14 , wherein said processing application is a VXML application, the grammar ID identifies a particular grammar and tag types include link, embedded prompt, and non-embedded prompt.
16. A telecommunications system in accordance with claim 15 , wherein said data identifies an audio file to be played.
17. A telecommunications system in accordance with claim 15 , wherein said data identifies an URL to be processed.
18. A telecommunications system in accordance with claim 15 , wherein said data identifies a prompt to be provided.
19. A telecommunications system in accordance with claim 15 , wherein the grammar is an XP grammar recognizing utterances that include multiple pieces of information.