PROVIDING SPEECH RECOGNITION DATA TO A SPEECH ENABLED DEVICE WHEN PROVIDING A NEW ENTRY THAT IS SELECTABLE VIA A SPEECH RECOGNITION INTERFACE OF THE DEVICE

• US 20090157392A1
• Filed: 12/18/2007
• Published: 06/18/2009
• Est. Priority Date: 12/18/2007
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: ALEWINE, Neal J., BADT, Daniel E.
• Granted Patent: US 8,010,345 B2
• Time in Patent Office: Days
• US Class Current: 704/201
• CPC Class Codes: G10L 15/06 Creation of reference templates; Training of speech recognition systems, e.g. adaptation to the characteristics of the speaker's voice G10L15/14 takes precedence
• CPC Class Codes: G10L 15/183 using context dependencies, e.g. language models

Abstract

The present invention discloses a solution for providing a phonetic representation for a content item along with a content item delivered to a speech enabled computing device. The phonetic representation can be specified in a manner that enables it to be added to a speech recognition grammar of the speech enabled computing device. Thus, the device can recognize speech commands using the newly added phonetic representation that involve the content item. Current implementations of speech recognition systems of this type rely internal generation of speech recognition data that is added to the speech recognition grammar. Generation of speech recognition data can, however, be resource intensive, which can be particularly problematic when the speech enabled device is resource limited. The disclosed solution offloads the task of providing the speech recognition data to an external device, such as a relatively resource rich server or a desktop device.

Claims

1. A method for offloading a task of generating speech recognition data for a recognition grammar comprising:

identifying at least one content item, which lacks an entry in a speech recognition grammar used by a speech enabled device;

a computing device external to the speech enabled device generating speech recognition data for the at least one content item;

conveying the generated speech recognition data in a digitally encoded format within a carrier wave to the speech enabled device; and

adding the generated speech recognition data to the speech recognition grammar, which permits the speech enabled device to identify speech input as being associated with the at least one content item.

2. The method of claim 1, wherein the computing device is a Web server, said method further comprising:

connecting the speech enabled device to the Web server over a network through which the generated speech recognition data is conveyed to the speech enabled device.

3. The method of claim 1, wherein the source is a desktop computer, and wherein said speech enabled device is a portable computing device, said method further comprising:

connecting the speech enabled device to the desktop computer using a direct communication link, which is at least one of a serial link and a radio frequency link, wherein said generated speech recognition data is conveyed to the speech enabled device over the direct communication link.

4. The method of claim 1, wherein the content item is a digitally encoded song, wherein the speech enabled device is a digital music player, wherein the external computing device is at least one of a computer connected to the digital music player via a direct communication link and a Web server connected to the digital music player over a network, said method further comprising:

receiving a user selection of the digitally encoded song; and

not receiving an explicit user selection for speech recognition data associated with the digitally encoded song, which is nevertheless generated by the external computing device and conveyed to the speech enabled device along with the explicitly requested digitally encoded song.

5. The method of claim 1, wherein said steps of claim 1 are performed by at least one machine in accordance with at least one computer program stored in a computer readable media, said computer programming having a plurality of code sections that are executable by the at least one machine.

6. A method for integrating new content into a speech enabled device comprising:

requesting at least one content item from a source external to a speech enabled device;

receiving the requested content item along with speech recognition data associated with the content item;

adding the received speech recognition data to a speech recognition grammar of the speech enabled device; and

adding the content item to a data store of the speech enabled device.

7. The method of claim 6, further comprising:

receiving speech input;

speech recognizing the received speech input using the speech recognition grammar, wherein results from the speech recognizing step are derived from the added speech recognition data and indicate that an operation related to the content item is to be performed; and

executing programmatic action involving the content item.

8. The method of claim 7, wherein the speech enabled device is a music playing device, and wherein the content item is a digitally encoded song.

9. The method of claim 6, further comprising:

receiving a user selection of the content item, where the requesting step is performed responsive to the receiving of the user selection, wherein said user selection explicitly requests the content item but fails to explicitly request the speech recognition data for the content item, which is provided automatically by a content source remote from the speech enabled device along with the content item.

10. The method of claim 6, wherein the source is a Web server, said method further comprising:

connecting the speech enabled device to the Web server over a network;

presenting a set of available content of a Web site managed by the Web server via an interface of the speech enabled device to a user of the speech enabled device;

receiving user input of a selection of the set of content, wherein the selection comprises the at least one content item; and

responsive to the user input, the speech enabled device performing the requesting step.

11. The method of claim 6, wherein the source is a desktop computer, and wherein said speech enabled device is a portable computing device, said method further comprising:

connecting the speech enabled device to the desktop computer using a direct communication link, which is at least one of a serial link and a radio frequency link.

12. The method of claim 6, wherein a computing device remote from the speech enabled device compiles the speech recognition grammar containing the added speech recognition data, wherein the compiled speech recognition grammar is conveyed in a digitally encoded form from the remote computing device to the speech enabled device, and wherein the speech enabled computing device thereafter stores the compiled speech recognition grammar in a local data store and utilizes the compiled speech recognition grammar for speech recognition purposes.

13. A method of providing content to a speech enabled device along with associated speech recognition data comprising:

receiving a request for a content item from a remotely located speech enabled device;

determining an identifier for the content item;

identifying speech recognition data representing a recognition grammar entry for the identifier; and

conveying the content item and the identified speech recognition data to the speech enabled device.

14. The method of claim 13, further comprising:

querying a data store containing a plurality of identifiers and associated speech recognition data entries; and

receiving said speech recognition data from the data store responsive to the querying step.

15. The method of claim 13, further comprising:

compiling a speech recognition grammar comprising the identified speech recognition data, wherein said conveying step conveys the content item and said compiled speech recognition grammar to the speech enabled device.

16. The method of claim 13, further comprising:

dynamically creating speech recognition data for the identifier, wherein the created speech recognition data is the speech recognition data of the identifying step; and

adding an entry to a data store external to the remotely located speech enabled device for the created speech recognition data and an associated entry for the identifier, wherein said added entry is utilized by the method to respond to future requests for the content item so that the dynamically creating step is unnecessary for these future requests.

17. The method of claim 13, further comprising:

identifying a set of device specific parameters for the speech enabled device; and

formatting the speech recognition data in accordance with a speech grammar specification standard compatible with the device specific parameters.

18. The method of claim 13, further comprising:

identifying a set of user specific parameters for a user of the speech enabled device, wherein said user specific parameters comprise speech characteristics of the associated user; and

customizing the speech recognition data during the dynamically creating step in accordance with the speech characteristics of the associated user.

19. The method of claim 13, wherein a Web server performs said steps of claim 13 in accordance with a set of programmatic instructions executed by the Web server that are stored in a machine readable medium, and wherein said speech enabled device is a client communicating with the Web server over a network, wherein said content item is one of a plurality of user selectable content items available for downloading to said client.

20. The method of claim 13, wherein said steps of claim 13 are performed by at least one machine in accordance with at least one computer program stored in a computer readable media, said computer programming having a plurality of code sections that are executable by the at least one machine.