Method and device for audio input routing

1. A method implemented by one or more processors of a device, the method comprising:

monitoring an audio input for an occurrence of a trigger phrase that is spoken by a user of the device while the device is in a low power mode;

in response to detecting the occurrence of the trigger phrase:

waking the device from the low power mode; and

buffering the audio input while the device wakes up from the low power mode; and

in response to the device waking up from the low power mode, monitoring the buffered audio input for an occurrence of a command phrase that is spoken by the user, wherein the command phrase corresponds to a call command and a call parameter;

in response to detecting the occurrence of the command phrase:

selecting a set of target contacts associated with the device based on respective voice validation scores and respective contact confidence scores, wherein the respective voice validation scores indicate a corresponding likelihood that the call parameter is associated with a corresponding target contact, included in the set of target contacts, and wherein the respective contact confidence scores indicate a corresponding additional likelihood that a user context associated with the user for the call action being directed to the corresponding target contact, included in the set of target contacts; and

in response determining that a given voice validation score, of the respective voice validation scores, for a priority contact, included in the set of target contacts, meets a validation threshold and in response to determining that a given contact confidence score, of the respective contact confidence scores, for the priority contact, included in the set of target contacts, meets a confidence threshold:

automatically placing a call to the priority contact.

2. The method of claim 1 , wherein selecting the set of target contacts comprises:

generating a set of voice validation scores that correspond to a set of call contacts associated with the device based on the command phrase; and

selecting the set of target contacts, from the set of call contacts, with a corresponding one of the respective voice validation scores within a selected range of a highest voice validation score of the set of voice validation scores.

3. The method of claim 2 , wherein generating the set of voice validation scores comprises:

providing the set of call contacts to a voice recognition engine;

providing the call parameter to the voice recognition engine; and

receiving the set of voice validation scores from the voice recognition engine.

4. The method of claim 2 , further comprising:

generating a set of contact confidence scores as the respective contact confidence scores, for each of the corresponding target contacts included in the set of target contacts, based on the user context associated with the user, wherein the user context comprises one or more of: a recency of last communication, frequency of communication, placement on a favorites list by the user, or relation to a calendar event or appointment for the user; and

selecting respective phone numbers for each of the corresponding target contacts included in the set of target contacts.

5. The method of claim 4 , wherein automatically placing the call to the priority contact comprises:

determining whether only a single contact of the set of target contacts has the given voice validation score that meets the validation threshold and the given contact confidence score that meets the confidence threshold; and

automatically placing the call to the phone number for the single contact as the priority contact.

6. The method of claim 4 , further comprising:

selecting a set of phone numbers that correspond to a single contact, included in the set of target contacts, if the set of target contacts has only the single contact as the priority contact;

automatically placing the call to a single phone number if:

the single contact has only the single phone number,

the given voice validation score of the single contact meets the validation threshold, and

the given contact confidence score of the single contact meets the confidence threshold; and

prompting the user to enter the single phone number if:

the single contact has only the single phone number but the given voice validation score of the single contact does not meet the validation threshold, or

the given contact confidence score of the single contact does not meet the confidence threshold.

7. The method of claim 4 , further comprising:

sorting the set of target contacts based on the set of contact confidence scores and the set of voice validation scores;

prompting the user to select the priority contact from the set of target contacts if the set of target contacts has more than one target contact.

8. The method of claim 1 , further comprising:

validating the call command based on an audio length of the command phrase.

9. A device comprising:

at least one processor; and

memory storing instructions that, when executed, cause the at least one processor of the device to:

monitor an audio input for an occurrence of a trigger phrase that is spoken by a user of the device while the device is in a low power mode;

in response to detecting the occurrence of the trigger phrase:

wake the device from the low power mode; and

buffer the audio input while the device wakes up from the low power mode; and

in response to the device waking up from the low power mode, monitor the buffered audio input for an occurrence of a command phrase that is spoken by the user, wherein the command phrase corresponds to a call command and a call parameter;

in response to detecting the occurrence of the command phrase:

select a set of target contacts associated with the device based on respective voice validation scores and respective contact confidence scores, wherein the respective voice validation scores indicate a corresponding likelihood that the call parameter is associated with a corresponding target contact, included in the set of target contacts, and wherein the respective contact confidence scores indicate a corresponding additional likelihood that a user context associated with the user for the call action being directed to the corresponding target contact, included in the set of target contacts; and

in response determining that a given voice validation score, of the respective voice validation scores, for a priority contact included in the set of target contacts, meets a validation threshold and in response to determining that a given contact confidence score, of the respective contact confidence scores, for the priority contact, included in the set of target contacts, meets a confidence threshold:

automatically place a call to the priority contact.

10. The device of claim 9 , wherein the instructions to select the set of target contacts comprise instructions to:

generate a set of voice validation scores that correspond to a set of call contacts associated with the device based on the command phrase; and

select the set of target contacts, from the set of call contacts, with a corresponding one of the respective voice validation scores within a selected range of a highest voice validation score of the set of voice validation scores.

11. The device of claim 10 , wherein the instructions the generate the set of voice validation scores comprise instructions to:

provide the set of call contacts to a voice recognition engine;

provide the call parameter to the voice recognition engine; and

receive the set of voice validation scores from the voice recognition engine.

12. The device of claim 10 , wherein the instructions further cause the at least one processor of the device to:

generate a set of contact confidence scores as the respective contact confidence scores, for each of the corresponding, for each of the corresponding target contacts included in the set of target contacts, based on the user context associated with the user, wherein the user context comprises at least one of a call history or a favorite contact list; and

select respective phone numbers for each of the corresponding target contacts included in the set of target contacts.

13. The device of claim 12 , wherein the instructions to automatically place the call comprise instructions to:

determine whether only a single contact of the set of target contacts, has the given voice validation score that meets the validation threshold and the given contact confidence score that meets the confidence threshold; and

automatically place the call to the phone number for the single contact as the priority contact.

14. The device of claim 12 , wherein the instructions further cause the at least one processor of the device to:

select a set of phone numbers that correspond to a single contact, included in the set of target contacts, if the set of target contacts has only the single contact as the priority contact;

automatically place the call to a single phone number if:

the given single contact has only the single phone number,

the given voice validation score of the single contact meets the validation threshold, and

the contact confidence score of the single contact meets the confidence threshold; and

prompt the user to place the call to the single phone number if:

the single contact has only the single phone number but the given voice validation score of the single contact does not meet the validation threshold, or

the given contact confidence score of the single contact does not meet the confidence threshold.

15. The device of claim 12 , wherein the instructions further cause the at least one processor of the device to:

sort the set of target contacts based on the set of contact confidence scores and the set of voice validation scores; and

prompt the user to select the priority contact from the set of target contacts if the set of target contacts has more than one target contact.

16. The device of claim 9 , wherein the instructions further cause the at least one processor of the device to:

validate the call command based on an audio length of the command phrase.

17. A non-transitory computer-readable storage medium storing instructions that, when executed, cause at least one processor of a device to perform operations, the operations comprising:

monitoring an audio input for an occurrence of a trigger phrase that is spoken by a user of the device while the device is in a low power mode;

in response to detecting the occurrence of the trigger phrase:

waking the device from the low power mode; and

buffering the audio input while the device wakes up from the low power mode; and

in response to the device waking up from the low power mode, monitoring the buffered audio input for an occurrence of a command phrase that is spoken by the user, wherein the command phrase corresponds to a call command and a call parameter;

in response to detecting the occurrence of the command phrase:

selecting a set of target contacts associated with the device based on respective voice validation scores and respective contact confidence scores, wherein the respective voice validation scores indicate a corresponding likelihood that the call parameter is associated with a corresponding target contact, included in the set of target contacts, and wherein the respective contact confidence scores indicate a corresponding additional likelihood that a user context associated with the user for the call action being directed to the corresponding target contact, included in the set of target contacts; and

in response determining that a given voice validation score, of the respective voice validation scores, for a priority contact, included in the set of target contacts, meets a validation threshold and in response to determining that a given contact confidence score, of the respective contact confidence scores, for the priority contact, included in the set of target contacts, meets a confidence threshold:

automatically placing a call to the priority contact.

18. The method of claim 1 , wherein the user context comprises one or more of: a recency of last communication, frequency of communication, placement on a favorites list by the user, or relation to a calendar event or appointment for the user.

19. The method of claim 1 , wherein the user context comprises one or more of: placement on a favorites list by the user, or relation to a calendar event or appointment for the user.