TECHNIQUES FOR WAKE-UP WORD RECOGNITION AND RELATED SYSTEMS AND METHODS

• US 20190311715A1
• Filed: 06/15/2016
• Published: 10/10/2019
• Est. Priority Date: 06/15/2016
• Status: Active Grant

• Current Assignee: Cerence Operating Company (Cerence Inc.)
• Original Assignee: Nuance Communications, Inc. (Microsoft Corporation)
• Inventors: Pfeffinger, Meik, Matheja, Timo, Herbig, Tobias, Haulick, Tim
• Granted Patent: US 11,600,269 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• 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 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: G10L 2015/088 Word spotting
• CPC Class Codes: G10L 2015/223 Execution procedure of a spoken command

Abstract

A system for detection of at least one designated wake-up word for at least one speech-enabled application. The system comprises at least one microphone; and at least one computer hardware processor configured to perform: receiving an acoustic signal generated by the at least one microphone at least in part as a result of receiving an utterance spoken by a speaker; obtaining information indicative of the speaker'"'"'s identity; interpreting the acoustic signal at least in part by determining, using the information indicative of the speaker'"'"'s identity and automated speech recognition, whether the utterance spoken by the speaker includes the at least one designated wake-up word; and interacting with the speaker based, at least in part, on results of the interpreting.

Claims

1. A system for detecting at least one designated wake-up word for at least one speech-enabled application, the system comprising:

at least one microphone; and

at least one computer hardware processor configured to perform;

receiving an acoustic signal generated by the at least one microphone at least in part as a result of receiving an utterance spoken by a speaker;

obtaining information indicative of the speaker'"'"'s identity;

interpreting the acoustic signal at least in part by determining, using the information indicative of the speaker'"'"'s identity and automated speech recognition, whether the utterance spoken by the speaker includes the at least one designated wake-up word; and

interacting with the speaker based, at least in part, on results of the interpreting.

2. The system of claim 1, wherein interacting with the speaker comprises allowing the speaker to control the at least one speech-enabled application when the utterance spoken by the speaker is determined to include the at least one designated wake-up word.

3. The system of claim 1,

wherein the interpreting further comprises determining, using the information indicative of the speaker'"'"'s identity, whether the speaker is authorized to control the at least one speech-enabled application, and

wherein interacting with the speaker comprises allowing the speaker to control the at least one speech-enabled application when it is determined that the speaker is authorized to control the at least one speech-enabled application and not allowing the speaker to control the at least one speech-enabled application when it is determined that the speaker is not authorized to control the at least one speech-enabled application.

4. The system of claim 1, wherein obtaining information indicative of the speaker'"'"'s identity comprises obtaining the speaker'"'"'s identity at least in part by processing the acoustic signal.

5. The system of claim 4, wherein obtaining the speaker'"'"'s identity comprises:

obtaining speech characteristics from the acoustic signal;

comparing the obtained speech characteristics against stored speech characteristics for each of multiple speakers registered with the system.

6. The system of claim 4, wherein determining whether the utterance spoken by the speaker includes the at least one designated wake-up word comprises:

identifying one or more wake-up words associated with the speaker'"'"'s identity; and

using automated speech recognition to determine whether the utterance spoken by the speaker includes a wake-up word in the one or more wake-up words, wherein the automated speech recognition is performed using the one or more wake-up words associated with the speaker identity.

7. The system of claim 1, wherein obtaining information indicative of the speaker'"'"'s identity comprises determining a position of the speaker in an environment.

8. The system of claim 7, wherein the at least one computer hardware processor is further configured to perform:

determining, using the position of the speaker in the environment, whether the speaker is authorized to control the at least one speech-enabled application, and

wherein interacting with the speaker comprises allowing the speaker to control the at least one speech-enabled application when it is determined that the speaker is authorized to control the at least one speech-enabled application and not allowing the speaker to control the at least one speech-enabled application when it is determined that the speaker is not authorized to control the at least one speech-enabled application.

9. The system of claim 7, wherein the speaker is inside a vehicle, and wherein determining the position of the speaker is performed based, at least in part, on information gathered by at least one sensor in the vehicle, wherein the at least one sensor is different from the at least one microphone.

10. The system of claim 1, wherein the at least one microphone includes a plurality of microphones, and wherein determining the position of the speaker is performed by using acoustic signals generated by the plurality of microphones.

11. The system of claim 1, wherein the at least one microphone includes a first microphone and a second microphone, wherein the acoustic signal is generated by the first microphone, and wherein the at least one computer hardware processor is further configured to perform:

receiving a second acoustic signal generated by the second microphone at least in part as a result of receiving, concurrently with the first microphone receiving the utterance spoken by the speaker, a second utterance spoken by a second speaker;

obtaining information indicative of the second speaker'"'"'s identity;

interpreting the second acoustic signal at least in part by determining, using the information indicative of the second speaker'"'"'s identity and automated speech recognition, whether the second utterance spoken by the second speaker includes a second designated wake-up word for a second speech-enabled application; and

interacting with the second speaker based, at least in part, on results of the interpreting.

12. The system of claim 1, wherein the at least one microphone comprises a plurality of microphones installed in a respective plurality of acoustic zones inside of a vehicle, wherein each of the plurality of acoustic zones comprises a seating area for a passenger in the vehicle.

13. The system of claim 1, wherein interacting with the speaker comprises inferring, based at least in part on the information indicative of the speaker'"'"'s identity, at least one action to take when interacting with the speaker.

14. The system of claim 1, wherein the at least one designated wake-up word comprises a first designated wake-up word for a first speech-enabled application of the at least one speech-enabled application, and wherein the first designated wake-up word is specific to the speaker such that no other speaker can use the first designated wake-up word.

15. The system of claim 1,

wherein obtaining information indicative of the speaker'"'"'s identity comprises obtaining the speaker'"'"'s identity;

and

wherein determining whether the utterance spoken by the speaker includes the at least one designated wake-up word comprises;

accessing a list of wake-up words associated with the speaker'"'"'s identity; and

determining whether the utterance includes any wake-up word in the list of wake-up words associated with the speaker'"'"'s identity.

16. The system of claim 1, wherein determining, using automated speech recognition, whether the utterance spoken by the speaker includes the at least one designated wake-up word comprises:

compensating for interference received by the at least one microphone by using the information associated with the speaker'"'"'s identity.

17. The system of claim 1, further comprising at least one data store, wherein the at least one computer hardware processor is further configured to store the information about the speaker'"'"'s identity in the at least one data store.

18. The system of claim 17, wherein the at least one data store comprises a plurality of data records including a first data record, the first data record comprising information selected from the group consisting of an identity of a particular speaker, a position of the particular speaker in an environment, a list of one or more wake-up words associated with the particular speaker, a list of one or more speech-enabled applications that the particular speaker is allowed to control, a list of one or more speech-enabled applications that the particular speaker is not allowed to control, and information obtained from one or more sensors.

19. (canceled)

20. A method for detecting at least one designated wake-up word for at least one speech-enabled application, the method comprising:

using at least one computer hardware processor to perform;

receiving an acoustic signal generated by at least one microphone at least in part as a result of receiving an utterance spoken by a speaker;

obtaining information indicative of the speaker'"'"'s identity;

interpreting the acoustic signal at least in part by determining, using the information indicative of the speaker'"'"'s identity and automated speech recognition, whether the utterance spoken by the speaker includes the at least one designated wake-up word; and

interacting with the speaker based, at least in part, on results of the interpreting.

21-36. -36. (canceled)

37. At least one non-transitory computer-readable storage medium storing processor-executable instructions that, when executed by at least one computer hardware processor, cause the at least one computer hardware processor to perform a method for detecting at least one designated wake-up word for at least one speech-enabled application, the method comprising:

receiving an acoustic signal generated by at least one microphone at least in part as a result of receiving an utterance spoken by a speaker;

obtaining information indicative of the speaker'"'"'s identity;

interpreting the acoustic signal at least in part by determining, using the information indicative of the speaker'"'"'s identity and automated speech recognition, whether the utterance spoken by the speaker includes the at least one designated wake-up word; and

interacting with the speaker based, at least in part, on results of the interpreting.

38-53. -53. (canceled)