Monitoring system with speech recognition

• US 7,881,939 B2
• Filed: 05/31/2005
• Issued: 02/01/2011
• Est. Priority Date: 05/31/2005
• Status: Active Grant

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Tice, Lee D.
• Primary Examiner(s): Dorvil; Richemond
• Assistant Examiner(s): ORTIZ SANCHEZ, MICHAEL
• Application Number:
  US11/141,125
• Publication Number:
  US 20060285651A1
• Time in Patent Office: 2,072 Days
• Field of Search: 704/270, 340/539.22, 340/539.12, 340/539.11, 340/539.14
• US Class Current: 704/271
• CPC Class Codes: H04L 12/2818 from a device located outside both the home and the home network access arrangements H04L12/2856; for remote control or remote monitoring of applications H04L67/025; arrangements for transmitting signals characterised by the use of a wireless eletrical link G08C17/00; telephonic communication systems adapted for combination with remote control systems H04M11/007
• CPC Class Codes: H04M 11/007 with remote control systems remote control in general G08C

Abstract

A system for monitoring conditions associated with an individual in a region includes at least one speech input transducer and speech processing software coupled thereto. Results of the speech processing can initiate communications with a displaced communications device such as a telephone or a computer to provide a source of feedback.

Claims

1. A system comprising:

at least one sensor of a physiological condition of an individual;

a plurality of audio input transducers distributed throughout a residence of the individual;

at least one discrete, stationary background sound source within the residence with a first transducer of the plurality of transducers located at an output of the at least one discrete, stationary background sound source;

control circuitry coupled to the sensor and the transducers, the control circuitry has a communications port, the control circuitry responds to speech received from the transducers and carries out a speech recognition process thereon, and in accordance with the results thereof, initiates communications with a displaced location via the communications port; and

software executed by the control circuitry, the software compensates each of the plurality of input transducers to cancel background sounds from the at least one discrete, stationary background sound source and to null an acoustic signal from each of the plurality of input transducers based upon differences in phase and amplitude of the background sounds detected among each of the plurality of input transducers, the software executed by the control circuitry carrying out speech recognition of audio provided by the individual from the nulled acoustic signal provided by at least some of the plurality of input transducers, said nulling including detecting an acoustic signal from the at least one discrete, stationary background sound source through the first transducer and from each of at least two second transducers of the plurality of transducers, determining a phase difference between the acoustic signal detected by the first transducer and each of the at least two second transducers, adjusting a phase of the signal detected by the first transducer for each of the at least two second transducers based upon the phase difference of each of the respective second transducers, inverting each of the phase adjusted signals, adjusting an amplitude of the respective inverted signals and adding the respective phase and amplitude adjusted signals to the respective acoustic signal detected by the at least two second transducers to null the at least one discrete, stationary background sound source within the respective acoustic signals from the at least two second transducers.

2. A system as in claim 1 which includes an audio output transducer coupled to the control circuitry, the control circuitry provides at least one speech output to the audio output transducer.

3. A system as in claim 1 where the control circuitry provides one of a plurality of speech outputs to the audio output transducer.

4. A system as in claim 3 where the software selects at least one of the speech outputs in response to results of the speech recognition process.

5. A system as in claim 3 where the control circuitry couples speech signals from an audio input transducer of the plurality of transducers to the port for communications to the displaced location.

6. A system as in claim 5 where at least some communications received from the displaced location via the port are coupled by the control circuitry to the audio output transducer.

7. A system as in claim 1 where the control circuitry evaluates, at least in part, outputs from the at least one sensor.

8. A system as in claim 6 where the control circuitry evaluates, at least in part, outputs from the at least one sensor.

9. A system as in claim 1 which includes a sensor compensation function.

31. A system as in claim 1 which includes a plurality of audio input transducers locatable throughout an area to be monitored.

32. A system as in claim 31 in which at least one of the plurality of audio input transducers is located proximate a local source of ambient background sound where the at least one audio input transducer is compensated to minimize the ambient background sound generated from the source.

10. A monitoring system comprising:

at least a first sensor for monitoring a physiological condition of a person and associated sensor interface circuitry;

a plurality of acoustical input transducers for monitoring acoustical signals and associated transducer interface circuitry distributed throughout a residence of the person;

at least one discrete, stationary background noise source within the residence with a first transducer of the plurality of acoustic input transducers located at an output of the at least one discrete, stationary background noise source;

a device that receives at least first sensor data and which receives data from the transducers and respective interface circuitry, to carry out at least a speech recognition function, and responsive thereto, to initiate communications with a displaced unit; and

control software executed by the device, the software compensates each of the plurality of input transducers to cancel background sounds from the at least one discrete, stationary sound source and to null an acoustic signal from each of the plurality of input transducers based upon differences in phase and amplitude of the background sounds detected between the first input transducer and each of at least two other input transducers of the plurality of input transducers said canceling including inverting a signal received through the first input transducer of the plurality of input transducers for each of the at least two other input transducers, adjusting an amplitude and phase of the inverted signals based upon the respective detected difference in phase and amplitude and adding the respective inverted and adjusted signals to a signal from the respective at least two input transducers the software executed by the device carrying out speech recognition on audio provided by the person from the nulled acoustic signal provided by the at least two input transducers.

11. A system as in claim 10 where the speech recognition function includes recognizing at least one of spoken words or sounds indicative of a condition needing attention.

12. A system as in claim 11 including a plurality of acoustic output transducers to provide output audio information to a person regarding the recognition of a message by the speech recognition function.

13. A system as in claim 10 which includes circuitry to transfer to the remote unit at least in part a recording of speech of a person indicating or describing an ongoing condition.

14. A system as in claim 10 where the speech recognition function recognizes members of a plurality of condition describing words.

15. A system as in claim 10 where the remote unit comprises a bi-directional verbal communications device.

16. A system as in claim 10 which also includes an intercom system.

17. A system as in claim 10 which also includes a video system.

18. A system as in claim 10 which includes at least one audible output transducer.

19. A system as in claim 18 wherein the device, in combination with the audible output transducer, provide local audible feedback.

20. A system as in claim 19 which includes a plurality of acoustical input transducers coupled to the device.

21. A system as in claim 19 where at least some of the input transducers have associated therewith a respective audible output transducer.

22. A system as in claim 19 where the input transducers are coupled to the device by one of a wired, or a wireless link.

23. A system as in claim 19 where the device includes a programmable processor and associated voice recognition software.

24. A system as in claim 12 wherein the control software provides the output audio information.

25. A system as in claim 24 where the device includes a programmable processor and associated voice recognition software.

26. A system as in claim 10 which includes circuitry to compensate at least some of the acoustic input transducers.

27. A system as in claim 24 which includes a digital signal processor coupled to the control software.

28. A system as in claim 25 wherein the control software pre-stores background sound indicia.

29. A system as in claim 10 which includes circuitry to compensate at least some of the acoustic input transducers to reduce interference noise signals not associated with selected speech.

30. A system as in claim 29 where signals from at least some of the acoustic input transducers associated with the interference noise signals are coupled to the circuitry to compensate.

33. A method comprising:

providing a plurality of audio input transducers located in an area to be monitored;

providing at least one discrete, stationary sound source within the area with a first of the plurality of audio input transducers located at an output of the at least one discrete, stationary sound source;

detecting audio received from the area;

nulling simultaneous ambient background audio from the at least one discrete, stationary sound source within the detected audio of an acoustic signal from at least two other of the plurality of audio input transducers by detecting a difference in phase and amplitude between a signal from the first transducer and a respective signal from at least two other input transducers of the plurality of transducers, inverting a signal received through the first transducer, adjusting an amplitude and phase of the inverted signal based upon the detected differences in phase and amplitude for each of the at least two other input transducers and adding the respective inverted and adjusted signal to a respective signal from the at least two other transducers;

locally performing speech recognition on the detected audio with nulled ambient background audio from at least some of the plurality of input transducers to distinguish a resident'"'"'s speech from simultaneous ambient local background audio from the at least one discrete, stationary sound source;

selecting at least one speech output in response to the results of the speech recognition;

coupling speech signals from the at least some audio input transducer to a communications port;

transmitting a signal from the communications port to at least one remote location, the signal conveying information concerning the resident.

34. A method as in claim 33 which includes relaying a signal from the at least one remote location back to the communication port.

35. A method as in claim 34 which includes coupling at least some of the signal received from the remote location to an audio output transducer.

36. A method as in claim 33 which includes compensating the at least one audio input transducer.

37. A method as in claim 33 which includes evaluating, at least in part, outputs from at least one sensor.

38. A method as in claim 33 which includes storing the resident'"'"'s speech local to voice recognition software.

39. A method as in claim 33 which includes recognizing at least one spoken word or sound to be indicative of a condition needing attention.

40. A method as in claim 33 which includes recognizing members of a plurality of condition describing words.