IP Library Granted Patent US 9,646,610
Granted Patent B2
US 9,646,610 · App. 13/663,850 · Granted May 9, 2017

Method and apparatus for activating a particular wireless communication device to accept speech and/or voice commands using identification data consisting of speech, voice, image recognition

Inventor: Dusan Macho (Arlington Heights, IL)
Assignee: MOTOROLA SOLUTIONS, INC.
G10L15/22G06F1/3203G10L17/22G10L13/033G10L15/265G10L2015/088
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 9,646,610
App. No.
13/663,850
Granted
May 9, 2017
Kind
B2
Abstract

An apparatus, method, and computer program for initiating a word spotting algorithm ( 220 ) on one or more wireless communication devices in a first power mode to detect a keyword data sequence ( 224 ) embedded within a sampled audio signal ( 222 ). In response to detecting the keyword data sequence ( 226 ), the word spotting algorithm is terminated and a plurality of identification algorithms ( 230 ) consisting of speech, voice, image recognition and a predetermined isolation time criterion are initiated on the one or more wireless communication devices operating in a second power mode to detect the presence of identification data ( 240 ). If identification data is detected on a particular wireless communication device it is activated to accept speech and/or voice commands ( 242 ). On the other hand, if identification data is not detected, the plurality of identification algorithms are terminated, and the word spotting algorithm is reinitiated on the one or more wireless communication devices that are then operating in the first power mode ( 244 ).

Claims (54)

1. A method, comprising:

initiating a word spotting algorithm on a battery powered mobile wireless communication device in a first power mode to detect a keyword data sequence embedded within a sampled audio signal, the first power mode being adapted for low power consumption of the battery during the word spotting algorithm, and

in response to detecting the keyword data sequence, terminating the word spotting algorithm and initiating a plurality of identification algorithms on the mobile wireless communication device operating in a second power mode to detect for a presence of identification data, the second power mode being adapted for higher power consumption of the battery during the plurality of identification algorithms, wherein the plurality identification algorithms consist of:

(i) a speech recognition algorithm operating on the sampled audio signal to identify if an embedded numeric or alphanumeric identity code matches a predetermined numeric or an alphanumeric identity code;

(ii) a voice recognition algorithm operating on the sampled audio signal to identify that one or more embedded voice print data matches a predetermined voice print data;

(iii) an isolation algorithm operating on the sampled audio signal adapted to identify a predetermined degree of isolation between the keyword data sequence and subsequent utterances; and,

(iv) an image recognition algorithm operating on one or more images obtained from an image capture device and identifying one or more recognized images corresponding to one or more reference images;

wherein if the identification data is detected, the mobile wireless communication device is activated,

wherein if the identification data is not detected, the plurality of identification algorithms are terminated, and the word spotting algorithm is reinitiated on the battery operated mobile wireless communication device operating in the first power mode.

2. The method of claim 1 , wherein initiating the word spotting algorithm comprises:

chunking data received from an analog to digital convertor into data frames;

extracting features embedded in each frame of chunked data for phoneme recognition; and,

comparing recognized phonemes to a sequence of phonemes in a keyword data reference sequence.

3. The method of claim 2 , wherein the model is a whole word model with the whole word adapted to have variations based on state transitions from phonemes prior to the whole word and after the whole word.

4. The method of claim 1 , wherein the first power mode is a low power operational mode adapted for reduced power consumption of the battery by allowing the battery powered mobile wireless communication device to continually receive acoustic signals from an open microphone to spot keywords without significantly shortening, operational battery life of the battery powered mobile wireless communications device.

5. The method of claim 1 , wherein the speech recognition algorithm further comprises:

employing a digit recognizer to identify if the embedded numeric or alphanumeric identity code matches the predetermined numeric or alphanumeric identity code.

6. The method of claim 1 , wherein the voice recognition algorithm is adapted for applying Mel-frequency cepstral coefficients to the sampled audio signal.

7. The method of claim 1 , wherein the first power mode being adapted for power consumption of the battery during the word spotting algorithm does not significantly impact operational battery life of the battery powered mobile wireless communications device.

8. The method of claim 1 , wherein the initiating of the word spotting algorithm is initiated without use of a mechanical activation button or software emulated activation button.

9. An apparatus, comprising:

at least one processor; and at least one tangible memory storing computer instructions that when executed, cause the apparatus to:

initiate a word spotting algorithm on a battery powered mobile wireless communication device in a first power mode to detect a keyword data sequence embedded within a sampled audio signal, the first power mode being adapted for low power consumption of the battery during the word spotting algorithm, and in response to detecting the keyword data sequence, terminating the word spotting algorithm and initiating a plurality of identification algorithms on the battery powered mobile wireless communication device operating in a second power mode to detect the presence of identification data, the second power mode being adapted for higher power consumption of the battery during the plurality of identification algorithms, wherein the identification algorithms consist of:

(i) a speech recognition algorithm operating on the sampled audio signal to identify if an embedded numeric or alphanumeric identity code matches a predetermined numeric or alphanumeric identity code;

(ii) a voice recognition algorithm operating on the sampled audio signal to identify that one or more embedded voice print data matches a predetermined voice print data;

(iii) an isolation algorithm operating on the sampled audio signal adapted to identify a predetermined degree of isolation between the keyword data sequence and subsequent utterances; and,

(iv) an image recognition algorithm operating on one or more images obtained from an image capture device and identifying one or more recognized images corresponding to one or more reference images;

wherein if the identification data is detected, the mobile wireless communication device is activated,

wherein if the identification data is not detected, the identification algorithm is terminated, and the word spotting algorithm is reinitiated on the battery powered mobile wireless communication device operating in the first power mode.

10. The apparatus of claim 9 , wherein the first power mode is a low power operational mode adapted for allowing the battery powered mobile wireless device to continually receive acoustic signals from a microphone.

11. The apparatus of claim 9 , wherein initiating the word spotting algorithm comprises:

chunking data received from an analogue to digital convertor into data frames;

extracting features embedded in each frame of chunked data for phoneme recognition:

comparing recognized phonemes to a sequence of phonemes in a keyword data reference sequence.

12. The apparatus of claim 9 , wherein the wireless communication device comprises a device selected from a group consisting of:

a cellular phone, a personal digital assistant, a wireless modem, and a mobile radio, and a laptop computer.

13. The apparatus of claim 9 , wherein the battery powered mobile wireless communication device comprises a wireless land mobile radio operating within a land mobile radio system (LMRS) of a public safety network.

14. The apparatus of claim 9 , wherein the first power mode being adapted for power consumption of the battery during the word spotting algorithm does not significantly impact operational battery life of the battery powered mobile wireless communications device.

15. The apparatus of claim 9 , wherein word spotting algorithm is initiated without use of a mechanical activation button or software emulated activation button.

16. A non-transitory program storage device readable by a machine, tangibly embodying a program of instructions executable by the machine for causing operations, said operations comprising:

initiating a word spotting algorithm on a battery powered mobile wireless communication device in a first power mode to detect a keyword data sequence embedded within a sampled audio signal, the first power mode being adapted for low power consumption of the battery during the word spotting algorithm, and in response to detecting the keyword data sequence, terminating the word spotting algorithm and initiating a plurality of identification algorithms on the battery powered mobile wireless communication device operating in a second power mode to detect the presence of identification data, the second power mode being adapted for higher power consumption of the battery during the plurality of identification algorithms, wherein the plurality of identification algorithms comprise:

(i) a speech recognition algorithm operating on the sampled audio signal to identify if an embedded numeric or alphanumeric identity code matches a predetermined numeric or alphanumeric identity code;

(ii) a voice recognition algorithm operating on the sampled audio signal to identify that one or more embedded voice print data matches a predetermined voice print data;

(iii) an isolation algorithm operating on the sampled audio signal adapted to identify a predetermined degree of isolation between the keyword data sequence and subsequent utterances; and,

(iv) an image recognition algorithm operating on one or more images obtained from an image capture device and identifying one or more recognized images corresponding to one or more reference images;

wherein if the identification data is detected, the mobile wireless communication device is activated,

wherein if the identification data is not detected, the one or more identification algorithms are terminated, and the word spotting algorithm is reinitiated on the battery powered mobile wireless communication device operating in the first power mode.

17. The non-transitory program storage device of claim 16 , wherein initiating the word spotting algorithm comprises:

chunking data received from an analogue to digital convertor into data frames;

extracting features embedded in each frame of chunked data for phoneme recognition;

comparing a sequence of recognized phonemes to a sequence of phonemes in a keyword data reference sequence.

18. The non-transitory program storage device of claim 16 , wherein the second power mode is a higher power operational mode than the first power mode.

19. The non-transitory program storage device of claim 16 , wherein the first power mode being adapted for power consumption of the battery during the word spotting algorithm does not significantly impact operational battery life of the battery powered mobile wireless communications device.

20. The non-transitory program storage device of claim 16 , wherein the initiating of the word spotting algorithm is initiated without use of a mechanical activation button or software emulated activation button.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Oct 30, 2012
From: MACHO, DUSAN
To: MOTOROLA SOLUTIONS, INC.
Reel/Frame 029211/0851 →
Continuity (1)
Related Publication 20140122087A1 · May 1, 2014