VOICEXML LANGUAGE EXTENSION FOR NATIVELY SUPPORTING VOICE ENROLLED GRAMMARS

• US 20080082963A1
• Filed: 10/02/2006
• Published: 04/03/2008
• Est. Priority Date: 10/02/2006
• Status: Active Grant

• Current Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Original Assignee: International Business Machines Corporation
• Inventors: MUSCHETT, BRIEN H.
• Granted Patent: US 7,881,932 B2
• Time in Patent Office: Days
• US Class Current: 717/114
• CPC Class Codes: G06F 40/143 Markup, e.g. Standard Generalized Markup Language [SGML] or Document Type Definition [DTD]
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/0631 Creating reference templates; Clustering
• CPC Class Codes: G10L 2015/228 of application context

Abstract

The present invention extends the VoiceXML language model to natively support voice enrolled grammars. Specifically, three VoiceXML tags can be added to the language model to add, modify, and delete acoustically provided phrases to voice enrolled grammars. Once created, the voice enrolled grammars can be used in normal speaker dependent speech recognition operations. That is, the voice enrolled grammars can be referenced and utilized just like text enrolled grammars can be referenced and utilized. For example using the present invention, voice enrolled grammars can be referenced by standard text-based Speech Recognition Grammar Specification (SRGS) grammars to create more complex, usable grammars.

Claims

1. A VoiceXML language comprising:

a plurality of voice enrolled grammar language elements of a VoiceXML language configured to natively support voice enrolled grammars.

2. The language of claim 1, wherein said language elements are configured to permit voice enrolled grammars created and maintained using the language elements to be referenced by standard text-based grammars.

3. The language of claim 2, wherein said text-based grammars are Speech Recognition Grammar Specification (SRGS) compliant grammars.

4. The language of claim 1, wherein said language elements comprise:

an addvoicephrase language element that is a VoiceXML element configured to add a voicephrase to a voice enrolled grammar; and

a deletevoicephrase element that is a VoiceXML element configured to delete a voice phrase from an existing voice enrolled grammar.

5. The language of claim 4, wherein said language elements further comprise:

an updatevoicephrase element that is a VoiceXML element configured to update a voice phrase included in an existing voice enrolled grammar.

6. The language of claim 5, wherein attribute values for the plurality of language elements conform to naming conventions specified for a language model of the VoiceXML language, and wherein phrase related values for voice enrolled grammars manipulated by the language elements conform to naming rules specified by a Speech Recognition Grammar Specification (SRGS) compliant grammar.

7. The language of claim 4, wherein the addvoicephrase language element comprises a plurality of attributes, said plurality of attributes comprising at least three attributes selected from a group of attributes consisting of a name attribute, a phraseid attribute, a phraseidexpr attribute, a src attribute, a srcexpr attribute, a phrasenl attribute, a phrasenlexpr attribute, a weight attribute, a recordphraseaudio attribute, a consistencylevel attribute, a classlevel attribute, a maxattempts attribute, a minconsistent attribute, a badphrasesuri attribute, and an audioiputurilist attribute.

8. The language of claim 4, wherein the addvoicephrase language element is configured to return values depending on an outcome of an attempt to enroll a voicephrase using the addvoicephrase language element, said return values comprising at least two returns selected from a group of returns consisting of a phraseid return, a maxattempts return, an audio.ref return, an audio.duration return, and an audio.size return.

9. The language of claim 4, wherein the addvoicephrase language element is configured to selectively generate an error.enroll.clash event and an error.enroll.maxattempts event, wherein the error.enroll.class event is generated when a clash is detected between a voice phrase attempted to be added using the addvoicephrase language element and at least one of an existing voice phrase in a cooresponding voice enrolled grammar and a voice phase contained within a list of bad voice phrases maintained for the cooresponding voice enrolled grammar, and wherein the error.enroll.maxattempts event is generated when consistency of a voice phrase attempted to be added using the addvoicephrase language element is not achieved within an established maximum number of attempts.

10. The language of claim 4, wherein the deletevoicephrase language element comprises a plurality of attributes, said plurality of attributes comprising at least two attributes selected from a group of attributes consisting of a phraseid attribute, a phraseidexpr attribute, a src attribute, and a srcexpr attribute.

11. The language of claim 4, wherein the deletevoicephrase language element is configured to selectively generate an error.enroll.deletevoicephrase event that is generated when an attempt to delete a voice phrase is made using the deletevoicephrase language element and when the voice phrase is not able to be deleted.

12. The language of claim 5, wherein the updatevoicephrase language element comprises a plurality of attributes, said plurality of attributes comprising at least two attributes selected from a group of attributes consisting of a phraseid attribute, a phraseidexpr attribute, a newphraseid attribute, a newphraseidexpr attribute, a src attribute, a srcexpr attribute, a phrasenl attribute, a phrasenlexpr attribute, and a weight attribute.

13. The language of claim 5, wherein the updatevoicephrase language element is configured to selectively generate an error.enroll.updatevoicephrase event that is generated when an attempt to update a voice phrase is made using the updatevoicephrase language element and when the voice phrase is not able to be updated.

14. Voice XML code for manipulating voice phrases contained within a voice enrolled grammar comprising:

a VoiceXML native tag selected from a list of tags consisting of an addvoicephrase tag, a deletevoicephrase tag, and an updatevoicephrase tag; and

an identifier associated with the VoiceXML native tag that identifies a voice phrase, wherein execution of the VoiceXML code results in a change to the voice enrolled grammar relating to the identified voice phrase, wherein specifics of the change are determined by specifics associated with the VoiceXML native tag.

15. The VoiceXML code of claim 14, wherein said VoiceXML native tag is an addvoicephrase tag, wherein the change adds the identified voice phrase to the voice enrolled grammar.

16. The VoiceXML code of claim 14, wherein the addvoicephrase tag is associated with a plurality of attributes, said plurality of attributes comprising at least three attributes elected from a group of attributes consisting of a name attribute, a phraseid attribute, a phraseidexpr attribute, a src attribute, a srcexpr attribute, a phrasenl attribute, a phrasenlexpr attribute, a weight attribute, a recordphraseaudio attribute, a consistencylevel attribute, a classlevel attribute, a maxattempts attribute, a minconsistent attribute, a badphrasesuri attribute, and an audioiputurilist attribute.

17. The VoiceXML code of claim 14, wherein said VoiceXML native tag is a deletevoicephrase tag, wherein the change deletes the identified voice phrase from the voice enrolled grammar.

18. The VoiceXML code of claim 17, wherein the deletevoicephrase tag is associated with a plurality of attributes, said plurality of attributes comprising at least two attributes selected from a group of attributes consisting of a phraseid attribute, a phraseidexpr attribute, a src attribute, and a srcexpr attribute.

19. The VoiceXML code of claim 14, wherein said VoiceXML native tag is an updatevoicephrase tag, wherein the change updates the identified voice phrase contained in the voice enrolled grammar.

20. The VoiceXML code of claim 19, wherein the updatevoicephrase tag is associated with a plurality of attributes, said plurality of attributes comprising at least three attributes selected from a group of attributes consisting of a phraseid attribute, a phraseidexpr attribute, a newphraseid attribute, a newphraseidexpr attribute, a src attribute, a srcexpr attribute, a phrasenl attribute, a phrasenlexpr attribute, and a weight attribute.