Methods and apparatus for unsupervised wakeup
View Patent ↗Methods and apparatus for unsupervised wakeup of a device including receiving a first acoustic event at a first time and a second acoustic event at a second time, wherein scores of the first and second acoustic events are above a first threshold identifying the first and second acoustic events as wakeup candidates for a wakeup phrase for an unsupervised wakeup of a device. It is determined that the first acoustic event is above a second threshold, which is higher than the first threshold, and that the second acoustic event is above a third threshold, which is higher than the first threshold. Occurrence of a wakeup event can be determined based upon acoustic similarity of the events.
1. A method, comprising:
receiving a first acoustic event at a first time and a second acoustic event at a second time from a microphone of a device, wherein scores of the first and second acoustic events are above a first threshold identifying the first and second acoustic events as wakeup candidates for a wakeup phrase for an unsupervised wakeup of the device;
determining, using a computer processor of the device, that the first acoustic event is above a second threshold, which is higher than the first threshold;
determining that the second acoustic event is above a third threshold, which is higher than the first threshold;
determining whether the first and second acoustic events are acoustically similar; and
determining that a wakeup event has occurred based upon the determination of acoustic similarity and performing a wakeup of the device.
2. The method according to claim 1 , further including adapting an acoustic model with the first and/or second acoustic event.
3. The method according to claim 1 , further including adapting an acoustic model with a lower scoring one of the first and second acoustic events.
4. The method according to claim 1 , wherein the second and third thresholds are the same.
5. The method according to claim 1 , further including using a main processor and a lower power processor to provide the unsupervised wakeup.
6. The method according to claim 5 , wherein the lower power processor identifies the first wakeup candidate.
7. The method according to claim 1 , further including receiving a textual input of the wakeup phrase.
8. The method according to claim 1 , further including receiving at least a third acoustic event above the second threshold, and determining that the at least a third acoustic event is acoustically similar to the first and/or second acoustic events.
9. An article, comprising:
a non-transitory computer readable medium having stored instructions that enable a machine to:
receive a first acoustic event at a first time and a second acoustic event at a second time from a microphone of a device, wherein scores of the first and second acoustic events are above a first threshold identifying the first and second acoustic events as wakeup candidates for a wakeup phrase for an unsupervised wakeup of a device;
determine, using a processor of the device, that the first acoustic event is above a second threshold, which is higher than the first threshold;
determine that the second acoustic event is above a third threshold, which is higher than the first threshold;
determine whether the first and second acoustic events are acoustically similar; and
determine that a wakeup event has occurred based upon the determination of acoustic similarity and perform a wakeup of the device.
10. The article according to claim 9 , further including instructions to adapt an acoustic model with the first and/or second acoustic event.
11. The article according to claim 9 , further including instructions to adapt an acoustic model with a lower scoring one of the first and second acoustic events.
12. The article according to claim 9 , wherein the second and third thresholds are the same.
13. The article according to claim 9 , further including instructions to use a main processor and a lower power processor to provide the unsupervised wakeup.
14. The article according to claim 13 , wherein the lower power processor identifies the first wakeup candidate.
15. The article according to claim 9 , further including instructions to receive at least a third acoustic event above the second threshold, and determine that the at least a third acoustic event is acoustically similar to the first and/or second acoustic events.
16. A system, comprising:
a processor and a memory configured to:
receive a first acoustic event at a first time and a second acoustic event at a second time from a microphone of a device, wherein scores of the first and second acoustic events are above a first threshold identifying the first and second acoustic events as wakeup candidates for a wakeup phrase for an unsupervised wakeup of a device;
determine, using the processor, that the first acoustic event is above a second threshold, which is higher than the first threshold;
determine that the second acoustic event is above a third threshold, which is higher than the first threshold;
determine whether the first and second acoustic events are acoustically similar; and
determine that a wakeup event has occurred based upon the determination of acoustic similarity and perform a wakeup of the device.
17. The system according to claim 16 , wherein the processor and memory are further configured to adapt an acoustic model with the first and/or second acoustic event.
18. The system according to claim 15 , wherein the second and third thresholds are the same.
19. The system according to claim 15 , wherein the processor and memory are further configured to use a main processor and a lower power processor to provide the unsupervised wakeup.
20. The system according to claim 15 , wherein the processor and memory are further configured to receive a textual input of the wakeup phrase.