Systems and methods of performing speech recognition using sensory inputs of human position

• US 20090094032A1
• Filed: 10/05/2007
• Published: 04/09/2009
• Est. Priority Date: 10/05/2007
• Status: Active Grant

• Current Assignee: Sensory Incorporated
• Original Assignee: Sensory Incorporated
• Inventors: Mozer, Todd F.
• Granted Patent: US 8,036,901 B2
• Time in Patent Office: Days
• US Class Current: 704/251
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 2015/227 of the speaker; Human-factor methodology

Abstract

Embodiments of the present invention improve methods of performing speech recognition using sensory inputs of human position. In one embodiment, the present invention includes a speech recognition method comprising sensing a change in position of at least one part of a human body, selecting a recognition set based on the change of position, receiving a speech input signal, and recognizing the speech input signal in the context of the first recognition set.

Claims

1. A speech recognition method comprising:

sensing a change of position of at least one part of a human body;

selecting a first recognition set based on the change of position;

receiving a speech input signal; and

recognizing the speech input signal in the context of the first recognition set, the recognizing resulting in a recognition result.

2. The method of claim 1 wherein the change of position includes a change in orientation.

3. The method of claim 1 wherein the change of position includes a change in direction.

4. The method of claim 1 wherein the change of position includes a portion of a human hand proximate with a surface.

5. The method of claim 1 wherein the change of position includes motion.

6. The method of claim 1 wherein the speech input signal is a portion of an utterance.

7. The method of claim 1 wherein the recognizing includes choosing an element from the first recognition set.

8. The method of claim 1 wherein the recognizing includes using the first recognition set to weight a set of likelihoods, wherein the first recognition set includes segments of speech.

9. The method of claim 1 further comprising:

initiating a state of a computer program; and

selecting a state recognition set based on the state of the computer program,

wherein the selecting of the first recognition set includes finding a subset of the state recognition set.

10. The method of claim 9 wherein the change of position includes a change in orientation.

11. The method of claim 9 wherein the change of position includes a change in a direction.

12. The method of claim 9 wherein the change of position includes a portion of the human hand proximate with a surface.

13. The method of claim 9 wherein the change of position includes motion.

14. The method of claim 9 wherein the recognizing includes choosing a select element from the first recognition set.

15. The method of claim 9 further comprising changing the state of the computer program according to the recognition result.

16. The method of claim 1 wherein the speech input signal includes digital data.

17. The method of claim 1 wherein the first recognition set includes segments of sound data.

18. The method of claim 1 wherein sensing includes the use of a tactile sensor.

19. The method of claim 1 wherein sensing includes the use of a gyroscope.