System and method for determining recipient of spoken command in a control system

Patent metadata

Field	Value
Publication number	US-10121473-B2
Application number	US-201615376846-A
Country	US
Kind code	B2
Filing date	Dec 13, 2016
Priority date	Sep 10, 2015
Publication date	Nov 6, 2018
Grant date	Nov 6, 2018

Abstract

Disclosed is an apparatus and method for determining which controllable device an audible command is directed towards, the method comprising: receiving at each of two or more controlling devices the audible command signal, the audible command being directed to control at least one of two or more controllable devices controlled by a respective one of the two or more controlling devices; digitizing each of the received audible command signals; attaching a unique identifier to each digitized audible command so as to uniquely correlate it to a respective controlling device; determining a magnitude of each of the digitized audible command; determining a digitized audible command with the greatest magnitude, and further determining to which controlling device the audible command is directed to on the basis of the unique identifier associated with the digitized audible command with the greatest magnitude; performing speech recognition on the digitized audible command with the greatest magnitude; and forwarding a command to the controlling device corresponding to the digitized audible command with the greatest magnitude, the command corresponding to the audible command that can be implemented on the controllable device controlled by the controlling device.

First claim

What is claimed is:

1. A method for determining which controllable device an audible command is directed towards, the method comprising:
receiving at each of two or more controlling devices the audible command signal, the audible command being directed to control at least one of two or more controllable devices controlled by a respective one of the two or more controlling devices;
digitizing each of the received audible command signals;
attaching a unique identifier to each digitized audible command so as to uniquely correlate it to a respective controlling device;
determining a magnitude of each of the digitized audible command;
determining a digitized audible command with the greatest magnitude, and further determining to which controlling device the audible command is directed to on the basis of the unique identifier associated with the digitized audible command with the greatest magnitude;
performing speech recognition on the digitized audible command with the greatest magnitude; and
forwarding a command to the controlling device corresponding to the digitized audible command with the greatest magnitude, the command corresponding to the audible command that can be implemented on the controllable device controlled by the controlling device, wherein the step of determining a magnitude further comprises—
attaching a time-date stamp to the digitized audible command; and
verifying that the time-date stamp of the greatest magnitude digitized audible command is the same or earlier than any other digitized audible command, and wherein the method further comprises—
applying additional processing to determine which controlling device the audible command is directed to if the time-date stamp of the digitized audible command with the largest magnitude is later than any other digitized audible command, and wherein

the step of applying additional processing comprises—
checking one or more of an occupancy sensor reading, proximity sensor reading, and motion detector reading, each of which is associated with respective controlling devices to determine which controlling device the audible command is directed to, and wherein the method further comprises—
determining that the occupancy sensor reading associated with a respective controlling device matches the audible command in order to determine that the audible command should be applied to the controllable device controlled by the respective controlling device.

2. The method according to claim 1 , further comprising:
receiving the command by the controlling device; and
controlling the controllable device in accordance with the received command.

3. The method according to claim 1 , wherein
the step of attaching a time-date stamp to the digitized audible command is performed by the controlling device that received the audible command through use of a time-date stamp generator using a time synch protocol.

4. The method according to claim 1 , wherein
the step of attaching a time-date stamp to the digitized audible command is performed by a central controller.

5. The method according to claim 4 , further comprising:
generating a test signal to determine a travel time from each of the plurality of controlling devices to the central controller; and
modifying the time-date stamp of each received digitized audible command signal according to the travel time from a respective controlling device to the central processor, and using the modified time-date stamp to assist in determining to which controlling device the audible command signal is directed to.

6. The method according to claim 1 , further comprising:
determining that the proximity sensor reading associated with a respective controlling device matches the audible command in order to determine that the audible command should be applied to the controllable device controlled by the respective controlling device.

7. The method according to claim 1 , further comprising:
determining that the motion detector reading associated with a respective controlling device matches the audible command in order to determine that the audible command should be applied to the controllable device controlled by the respective controlling device.

8. The method according to claim 1 , wherein the step of attaching a time-date stamp comprises:
generating the time-date stamp using a time-date stamp generator, the time-date stamp generator located in at least one controlling device that includes a plurality of microphones located at the controlling device;
averaging all of the time-date stamps generated at the controlling device with a plurality of microphones; and
attaching the averaged time-date stamp to each of the plurality of digitized audible commands prior to the step of transmitting.

9. The method according to claim 1 , further comprising:
applying noise reduction processing.

10. The method according to claim 9 , wherein the step of applying noise reduction to the audible command comprises:
filtering the received audible command signal in an analog domain to attenuate a first bandwidth of noise energy.

11. The method according to claim 9 , wherein the step of applying noise reduction comprises:
using one or more of acoustic echo cancellation filtering, direction of arrival filtering, and directionally adaptive beam forming filtering, to filter the digitized audible command in a digital domain to attenuate noise energy and to amplify audible command energy.

12. The method according to claim 11 , wherein the noise energy comprises:
noise energy generated by one or more of a fan motor, music, air conditioning noise, audio generated by multi-media presentations, and non-command words.

13. The method according to claim 1 , further comprising:
receiving at least one additional audible command from at least one additional spatially separated microphone, the at least one additional spatially separated microphone associated with a respective controlling device;
digitizing the at least one additional audible command received from the at least one additional spatially separated microphone;
attaching a time-date stamp to the at least one additional digitized audible command, and further attaching a unique identifier to each time-date stamped additional digitized audible command so as to uniquely correlate it to its respective controlling device; and
using the at least one additional digitized audible command in the step of determining to which controlling device the audible command is directed to.

14. The method according to claim 1 , wherein
the steps of determining a magnitude, determining to which controlling device the audible command is directed to, performing, and forwarding are performed by a central controller.

15. The method according to claim 1 , wherein
the steps of receiving, digitizing, attaching the time-date stamp, and determining a magnitude, are performed by the controlling device.

16. The method according to claim 1 , wherein
the controllable devices include one or more of a sensor, lighting control device, shade device, audio/video device, environmental control device, security device, household appliance, control device, and industrial device.

17. The method according to claim 1 , wherein the controlling device comprises:
a keypad.

18. An acoustic sensor network comprising:
two or more controllable devices;
two or more controlling devices, each of two or more controlling devices comprising at least one respective microphone, and an analog to digital converter, each of the two or more controlling devices being adapted to co

Assignees

• Crestron Electronics Inc

Inventors

• Bargetzi Fred
• Seferian Ara
• Stene Josh
• Labosco Mark

Classifications

• G10L15/22 Primary

  Procedures used during a speech recognition process, e.g. man-machine dialogue · CPC title

• G06F3/167

  Audio in a user interface, e.g. using voice commands for navigating, audio feedback · CPC title

• G10L21/0208

  Noise filtering · CPC title

• G10L2015/223

  Execution procedure of a spoken command · CPC title

• G10L15/30

  Distributed recognition, e.g. in client-server systems, for mobile phones or network applications · CPC title