Mechanism and apparatus for seamless voice wake and speaker verification

• US 9,445,209 B2
• Filed: 07/11/2013
• Issued: 09/13/2016
• Est. Priority Date: 07/11/2013
• Status: Active Grant

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Dadu, Saurabh, Krishnamurthy, Lakshman, Tharappel, Francis M., Datta, Prabhakar R., Peebler, Bryan R., Deisher, Michael E., Joshi, Mandar S., Shah, Saurin, Vogel, Brian K.
• Primary Examiner(s): AZAD, ABUL K
• Application Number:
  US14/129,644
• Publication Number:
  US 20150245154A1
• Time in Patent Office: 1,160 Days
• Field of Search: 704251-257, 704/270, 704/275, 455/404.1, 455/410, 455/411, 4553431-3436, 455/127.5
• US Class Current: 1/1
• CPC Class Codes: G06F 1/3206 Monitoring of events, devices or parameters that trigger a change in power modality
• CPC Class Codes: G06F 1/3231 Monitoring the presence, absence or movement of users
• CPC Class Codes: G06F 1/3287 by switching off individual functional units in the computer system
• CPC Class Codes: G06F 1/3293 by switching to a less power-consuming processor, e.g. sub-CPU
• CPC Class Codes: G06F 21/32 using biometric data, e.g. fingerprints, iris scans or voiceprints
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G10L 15/22 Procedures used during a speech recognition process, e.g. man-machine dialogue
• CPC Class Codes: G10L 15/285 Memory allocation or algorithm optimisation to reduce hardware requirements
• CPC Class Codes: G10L 17/22 Interactive procedures; Man-machine interfaces
• CPC Class Codes: G10L 19/0018 Speech coding using phonetic or linguistical decoding of the source; Reconstruction using text-to-speech synthesis
• CPC Class Codes: H04R 1/08 Mouthpieces; Microphones; Attachments therefor
• CPC Class Codes: H04R 29/004 for microphones H04R29/007 takes precedence
• CPC Class Codes: Y02D 10/00 Energy efficient computing, e.g. low power processors, power management or thermal management

Abstract

Technologies are described herein that allow a user to wake up a computing device operating in a low-power state and for the user to be verified by speaking a single wake phrase. Wake phrase recognition is performed by a low-power engine. In some embodiments, the low-power engine may also perform speaker verification. In other embodiments, the mobile device wakes up after a wake phrase is recognized and a component other than the low-power engine performs speaker verification on a portion of the audio input comprising the wake phrase. More than one wake phrases may be associated with a particular user, and separate users may be associated with different wake phrases. Different wake phrases may cause the device transition from a low-power state to various active states.

Claims

1. A computing device, comprising:

a microphone;

a low-power engine to;

recognize a wake phrase in audio input provided by a user to the microphone while the computing device is in a low-power state;

attempt to verify the user based at least in part on a portion of the audio input comprising the wake phrase while the computing device is in the low-power state; and

cause the computing device to transition from the low-power state to an active state after recognition of the wake phrase in the audio input provided by the user to the microphone while the computing device is in the low-power state and after the attempt to verify the user based at least in part on the portion of the audio input comprising the wake phrase while the computing device is in the low power state has failed; and

a command interpretation module to;

interpret a second portion of the audio input as a command to be executed by the computing device and to instruct the computing device to execute the command if the command is a user agnostic command, the second portion of the audio input not comprising the wake phrase.

2. The computing device of claim 1 wherein the low-power engine is to cause the computing device to transition from a second low-power state to a second active state after recognizing the wake phrase in a second audio input provided by a second user to the microphone while the computing device is in the second low-power state and after verifying the second user based at least in part on a portion of the second audio input comprising the wake phrase while the computing device is in the second low-power state.

3. The computing device of claim 2, wherein the command interpretation module is to interpret the portion of the second audio input comprising the wake phrase as a second command to be performed by the computing device.

4. The computing device of claim 1, wherein the active state is based at least in part on the wake phrase.

5. The computing device of claim 1, wherein the low-power engine is included in an audio capture pipeline of the computing device when the computing device is in the low-power state and the low-power engine is removed from the audio capture pipeline when the computing device is in the active state.

6. A method of waking a computing device, comprising:

while the computing device is in a low-power state;

receiving audio input from a user at the computing device;

determining that the audio input comprises a wake phrase; and

attempting to verify the user based at least in part on a portion of the audio input comprising the wake phrase;

transitioning the computing device from the low-power state to an active state if the audio input comprises the wake phrase and the user has not been verified based at least in part on the portion of the audio comprising the wake phrase;

interpreting a second portion of the audio input as a command to be executed by the computing device, the second portion of the audio input not comprising the wake phrase; and

performing the command at the computing device if the command is a user agnostic command.

7. The method of claim 6, wherein the wake phrase is one of a plurality of wake phrases that may be used to verify a plurality of users, the plurality of wake phrases comprising a first wake phrase that may be used to verify a first user and a second wake phrase that may be used to verify a second user, the first wake phrase being different from the second wake phrase.

8. The method of claim 6, wherein the determining that the audio input comprises the wake phrase is performed by a low-power engine of the computing device and the attempting to verify the user is performed by one or more components of the computing device other than the low-power engine.

9. The method of claim 6, further comprising:

interpreting a second audio input comprising the wake phrase as a second command to be executed by the computing device; and

performing the second command at the computing device.

10. One or more non-transitory computer-readable storage media storing computer-executable instructions for causing a computing device to:

while the computing device is in a low-power state;

receive audio input from a user at the computing device;

determine that the audio input comprises a wake phrase; and

attempt to verify the user based at least in part on a portion of the audio input comprising the wake phrase;

transition the computing device from the low-power state to an active state if the audio input comprises the wake phrase and the user has not been verified based at least in part on the portion of the audio comprising the wake phrase;

interpret a second portion of the audio input as a command to be executed by the computing device, the second portion of the audio input not comprising the wake phrase; and

perform the command at the computing device if the command is a user agnostic command.

11. The one or more computer-readable storage media of claim 10, wherein the wake phrase is one of a plurality of wake phrases that may be used to verify the user.

12. The one or more computer-readable storage media of claim 10, wherein the wake phrase is one of a plurality of wake phrases that may be used to verify a plurality of users, the plurality of wake phrases comprising a first wake phrase that may be used to verify a first user and a second wake phrase that may be used to verify a second user, the first wake phrase being different from the second wake phrase.

13. The one or more computer-readable storage media of claim 10, wherein the determining that the audio input comprises the wake phrase is performed by a low-power engine of the computing device and the attempting to verify the user is performed by one or more components of the computing device other than the low-power engine.

14. The one or more computer-readable storage media of claim 10, further comprising computer-executable instructions for causing the computing device to:

interpret a second audio input comprising the wake phrase as a second command to be executed by the computing device; and

perform the second command at the computing device.

15. The one or more computer-readable storage media of claim 10, wherein the transitioning comprises transitioning to the active state based at least in part on the wake phrase.

16. The one or more computer-readable storage media of claim 10, wherein a low-power engine is to cause the computing device to transition from a second low-power state to a second active state after recognizing the wake phrase in a second audio input provided by a second user to the microphone while the computing device is in the second low-power state and after verifying the second user based at least in part on a portion of the second audio input comprising the wake phrase while the computing device is in the second low-power state.