Dialog recognition and control in a voice browser

• US 7,003,464 B2
• Filed: 01/09/2003
• Issued: 02/21/2006
• Est. Priority Date: 01/09/2003
• Status: Active Grant

• Current Assignee: Google Technology Holdings LLC (Alphabet Inc.)
• Original Assignee: Motorola, Inc. (Motorola Solutions, Inc.)
• Inventors: Randolph, Mark, Vogedes, Jerome, Pearce, Michael, Ferrans, James, Engelsma, Jonathan
• Primary Examiner(s): CHAWAN, VIJAY B
• Application Number:
  US10/339,067
• Publication Number:
  US 20040138890A1
• Time in Patent Office: 1,139 Days
• Field of Search: 715/760, 704/270.1, 704/275, 704/200, 704/233, 704/277, 704/260, 704/270, 704/201, 704/246, 704/231
• US Class Current: 704/270.1
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: H04M 1/72445 for supporting Internet browser applications
• CPC Class Codes: H04M 2207/40 terminals with audio html browser
• CPC Class Codes: H04M 2250/74 with voice recognition means voice activated dialling H04M1/271; voice recognition algorithms G10L15/00
• CPC Class Codes: H04M 3/4938 comprising a voice browser which renders and interprets, e.g. VoiceXML

Abstract

A voice browser dialog enabler for multimodal dialog uses a multimodal markup document with fields have markup-based forms associated with each field and defining fragments. A voice browser driver resides on a communication device and provides the fragments and identifiers that identify the fragments. A voice browser implementation resides on a remote voice server and receives the fragments from the driver and downloads a plurality of speech grammars. Input speech is matched against those speech grammars associated with the corresponding identifiers received in a recognition request from the voice browser driver.

Claims

1. A voice browser dialog enabler for a communication system, the browser enabler comprising:

a speech recognition application comprising a plurality of units of application interaction, wherein each unit has associated voice dialog forms defining fragments;

a voice browser driver, the voice browser driver resident on a mobile communication device;

the voice browser driver providing the fragments from the application and generating identifiers that identify the fragments; and

a voice browser implementation resident on a remote voice server, the voice browser implementation receiving the fragments from the voice browser driver and downloading a plurality of speech grammars, wherein subsequent input speech is matched against those speech grammars associated with the corresponding identifiers received in a speech recognition request from the voice browser driver.

2. The voice browser enabler of claim 1, wherein the speech recognition request and subsequent speech recognition results are markup-based.

3. The voice browser enabler of claim 1, wherein the fragments consist of a VoiceXML page of identified forms.

4. The voice browser enabler of claim 3, wherein the fragments are cached by the voice browser implementation.

5. The voice browser enabler of claim 1, wherein the speech recognition application is a multimodal browser that processes multimodal markup documents, and the voice browser driver is resident in a voice browser stub that operates on a multimodal markup document to split the multimodal markup document into a displayable markup portion and a voice markup portion, and wherein the voice browser driver and voice browser implementation are operable on the voice markup portion.

6. The voice browser enabler of claim 5, further comprising an Internet application server with web server containing the multimodal markup document and the speech grammars.

7. The voice browser enabler of claim 5, further comprising a visual browser in the communication device that is operable on both the displayable markup portion and a voice markup portion of the multimodal markup document.

8. A voice browser for multimodal dialog in a communication system, the browser comprising:

a multimodal markup document split into a displayable markup portion and a voice markup portion comprising fields, wherein the fields have associated forms defining fragments of the document page;

a voice browser stub including a voice browser driver portion of a voice browser, the voice browser driver resident on a mobile communication device, the voice browser stub generating the fragments and the voice browser driver generating identifiers that identify the fragments; and

a voice browser implementation portion of the voice browser resident on a remote voice server, the voice browser implementation down loading the fragments from the voice browser stub and downloading a plurality of speech grammars, wherein subsequent input speech is matched against those speech grammars associated with the corresponding identifiers received in a speech recognition request from the voice browser driver.

9. The voice browser of claim 8, wherein the fragments are VoiceXML forms.

10. The voice browser of claim 8, wherein the speech recognition request and subsequent speech recognition results are markup-based.

11. The voice browser of claim 8, further comprising an Internet application server with web server containing the multimodal markup document and the speech grammars.

12. The voice browser of claim 8, further comprising a visual browser in the communication device that is operable on both the displayable markup portion and a voice markup portion of the multimodal markup document.

13. A method for enabling dialog with a voice browser for a communication system, the method comprising the steps of:

providing a voice browser driver resident on a communication device and a voice browser implementation containing a plurality of speech grammars resident on a remote voice serve;

running a speech recognition application comprising a plurality of units of application interaction, wherein each unit has associated voice dialog forms defining fragments;

defining identifiers associated with each fragment;

supplying the fragments to the voice browser implementation;

focusing on a field in one of the units of application interaction;

sending a speech recognition request including the identifier of the form associated with the focused field from the voice browser driver to the voice browser implementation;

inputting and recognizing speech;

matching the speech to the acceptable speech grammar associated with the identifier; and

obtaining speech recognition results.

14. The method of claim 13, wherein the speech recognition request of the sending step and the speech recognition results of the obtaining steps are markup-based.

15. The method of claim 13, wherein the supplying step includes supplying the voice browser implementation with a VoiceXML page of identified forms.

16. The method of claim 13, wherein the providing step includes the voice browser driver incorporated with a synchronizer into a voice browser stub that interfaces with the voice browser implementation and a visual browser on the communication device.

17. The method of claim 13, wherein the synchronizer is operable to enable an input from an Interact server when speech is detected.

18. The method of claim 13, wherein the running step includes downloading a multimodal markup document as the speech recognition application document.

19. The method of claim 18, wherein after the running step further comprising the step of splitting the multimodal markup document into a displayable markup portion and a voice markup portion containing die units of interaction, and wherein the subsequent steps are operable for only the voice markup portion of the document.

20. The method of claim 19, wherein the providing step includes providing a visual browser in the communication device that is operable on both the displayable markup portion and a voice markup portion of the multimodal markup document.