IDENTIFYING UN-STORED VOICE COMMANDS

• US 20160078864A1
• Filed: 09/15/2014
• Published: 03/17/2016
• Est. Priority Date: 09/15/2014
• Status: Abandoned Application

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Palanisamy, Prabhu, Mahasenan, Arun Vijayakumari
• Application Number:
  US14/486,786
• Publication Number:
  US 20160078864A1
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G10L 15/08 Speech classification or search
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/26 Speech to text systems G10L15/08 takes precedence
• CPC Class Codes: G10L 2015/221 Announcement of recognition results
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command

Abstract

Devices, methods, and computer-readable and executable instructions for identifying un-stored voice commands are described herein. For example, one or more embodiments include a microphone component configured to capture an un-stored voice command issued by a user and a speech recognition engine. The speech recognition engine can be configured to convert the un-stored voice command to device recognizable text, compare the device recognized text of the un-stored voice command to a plurality of stored voice commands of a voice controlled device, and identify a stored voice command among the plurality of stored voice commands based on the comparison of the device recognizable text of the un-stored voice command to the plurality of stored voice commands.

Claims

1. A voice controlled device, comprising:

a microphone component configured to capture an un-stored voice command issued by a user; and

a speech recognition engine configured to;

convert the un-stored voice command to device recognizable text;

compare the device recognized text of the un-stored voice command to a plurality of stored voice commands of a voice controlled device; and

identify a stored voice command among the plurality of stored voice commands based on the comparison of the device recognizable text of the un-stored voice command to the plurality of stored voice commands.

2. The device of claim 1, including a user interface to display the identified stored voice command.

3. The device of claim 1, wherein the identification of the stored voice command includes a prediction of the stored voice command the user intended by the un-stored voice command.

4. The device of claim 1, wherein the stored voice command identified includes a subset of the plurality of stored voice commands of the voice controlled device.

5. The device of claim 1, including an input component configured to receive an input from the user to add a new voice command to the plurality of stored voice commands.

6. The device of claim 1, wherein the identified stored voice command is a stored voice command among the plurality of stored voice commands that matches at least a portion of the device recognizable text.

7. The device of claim 6, wherein the match includes a keyword match of one or more words of the device recognizable text to the identified stored voice command in a stored voice commands file.

8. A non-transitory computer-readable medium, comprising instructions executable by a processing resource to cause a computing device to:

receive an un-stored voice command streamed from a microphone component of a voice controlled device;

convert the un-stored voice command to device recognizable text;

compare the device recognizable text of the un-stored voice command to a plurality of stored voice commands of the voice controlled device;

identify a stored voice command among the plurality of stored voice commands based on the comparison of the device recognizable text of the un-stored voice command to the plurality of stored voice commands; and

inform a user of the identified stored voice command via text-to-speech and/or a display on a screen.

9. The medium of claim 8, including instructions executable by the processing resource to identify the un-stored voice command is not one of the plurality of stored voice command by identification that the device recognizable text does not identically match any of the plurality of stored voice commands.

10. The medium of claim 8, wherein the stored voice commands are stored the computer-readable medium and include device specific commands executable by the processing resource to cause the computing device to perform functions.

11. The medium of claim 8, wherein the instructions are executable to:

identify a subset of the plurality of stored voice commands based on the comparison; and

provide a display of a list of sub-commands of the subset of the plurality of stored voice commands to inform the user of the subset.

12. The medium of claim 11, including instructions executable by the processing resource to receive a subsequent voice command from the user using the microphone component, wherein the subsequent voice command includes a selected one of the plurality of sub-commands in the list.

13. The medium of claim 12, including instructions executable by the processing resource to revise the list to include at least one of a sub-command of the selected sub-command and a stored voice command associated with the selected sub-command.

14. The medium of claim 12, including instructions executable by the processing resource to perform a function associated with the selected sub-command in response to user input.

15. A method for identifying an unknown voice command;

capturing a plurality of voice commands from a user using a microphone component of a voice controlled device;

streaming the plurality of captured voice commands to a speech recognition engine of the voice controlled device;

converting the plurality of captured voice commands to device recognizable text;

identifying at least a first voice command of the plurality of captured voice commands is an un-stored voice command based on the respective device recognizable text;

comparing the respective device recognizable text of the at least first voice command to a stored voice commands file;

identifying a subset of the plurality of stored voice commands based on the comparison of the respective device recognizable text to the stored voice commands file; and

informing a user of the subset of the plurality of stored voice commands.

16. The method of claim 15, including performing a stored voice command in the subset of the plurality of stored voice commands in response to user input.

17. The method of claim 16, wherein the user input includes a subsequent stored voice command issued by the user, and the method further includes:

capturing the subsequent stored voice command from the user using the microphone; and

converting the subsequent stored voice command to device recognizable text.

18. The method of claim 15, wherein informing the user of the subset includes:

providing a list of the subset of the plurality of stored voice commands, wherein each stored voice command in the subsets includes a sub-command of a stored voice command that is output from a keyword search of one or more words of the device recognizable test, and the method further includes;

revising the list to include sub-commands of a selected sub-command among the plurality of sub-commands.

19. The method of claim 15, including identifying at least a second voice command of the plurality of captured voice commands is a stored voice command among the plurality of stored voice commands.

20. The method of claim 19, including performing a function associated with the second voice command in response to identifying the second voice command is one of the plurality of stored voice command.