Device control based on signature
Systems, apparatuses, and methods are described for identifying a recurring event associated with a premises, and automatically performing actions associated with the recurring event. A computing device may generate a signature for the recurring event to indicate a series of actions associated with the recurring event, may detect occurrence of the recurring event, and may perform one or more of the series of actions indicated by the signature.
1. A method comprising:
receiving, by a computing device and during a time period associated with an entry into a premises by a user, audio data associated with sounds occurring during the entry, wherein the audio data comprises a first series of sound type indicators, and wherein the sound type indicators of the first series indicate types of the sounds occurring during the entry;
comparing the first series of sound type indicators to a second series of sound type indicators associated with one of a plurality of audio signatures, wherein the sound type indicators of the second series indicate multiple types of sounds, wherein each of the sound type indicators of the second series is associated with one of multiple time intervals during a signature period, and wherein the second series is associated with a sequence of actions performed by the user for each of a plurality of entries into the premises; and
causing, based on the comparing, a device associated with the premises to perform an action.
2. The method of claim 1 , wherein the one of the plurality of audio signatures comprises, for at least one of the multiple time intervals, multiple sound type indicators.
3. The method of claim 1 , further comprising:
updating, based on the received audio data, the one of the plurality of audio signatures.
4. The method of claim 1 , further comprising:
determining, based on one of the sound type indicators of the second series, the action.
5. The method of claim 1 , wherein:
the comparing comprises: comparing the first series of sound type indicators to an additional audio signature,
the one of the plurality of audio signatures is associated with a first recurring time window,
the additional audio signature is associated with a second recurring time window different than the first recurring time window, and
the additional audio signature comprises one or more sound type indicators also comprised by the one of the plurality of audio signatures.
6. The method of claim 1 , wherein the causing the device to perform the action comprises one or more of:
disarming a security system associated with the premises, or
causing a device associated with the premises to output content of a service associated with the one of the plurality of audio signatures.
7. The method of claim 1 , wherein the sound type indicators of the second series indicate one or more of the following types of sounds: a human voice sound type, an animal sounds sound type, an inaudible mechanical sounds sound type, an audible mechanical sounds sound type, an alert sounds sound type, or an absence of sound sound type.
8. The method of claim 1 , further comprising:
generating, based on audio data collected during a training period, the plurality of audio signatures.
9. The method of claim 1 , wherein the comparing comprises:
determining that a relationship, between a score for the received audio data and a score for the one of the plurality of audio signatures, satisfies a threshold.
10. The method of claim 9 , further comprising determining the score for the one of the plurality of audio signatures based on:
a quantity of the multiple types of sounds indicated by the second series of sound type indicators, and
a quantity of occurrences of each of the multiple types of sounds indicated by the second series of sound type indicators.
11. The method of claim 9 , further comprising determining the score for the received audio data based on:
a quantity of common types of sound, wherein the common types of sound are types of sound that are indicated by the first series of sound type indicators and by the second series of sound type indicators,
an order, indicated by the first series of sound type indicators, of the common types of sound, and
an order, indicated by the second series of sound type indicators, of the common types of sound.
12. The method of claim 1 , further comprising:
generating, based on audio data received during time periods associated with previous entries into the premises and comprising sound type indicators of the first series, the plurality of audio signatures.
13. The method of claim 1 , wherein at least one of the plurality of audio signatures comprises a sound type indicator that indicates a verbal sound type.
14. The method of claim 1 , wherein the multiple types of sounds comprise a sound associated with the user interacting with an object in the premises.
15. A method comprising:
determining, by a computing device, a score for one of a plurality of audio signatures, wherein each of the plurality of audio signatures is associated with a behavior pattern of a user, of a plurality of users, entering a premises, and wherein the score for the one of the plurality of audio signatures is determined based on:
a quantity of sound types indicated by the one of the plurality of audio signatures, and
a quantity of occurrences of each of the sound types indicated by the one of the plurality of audio signatures;
determining a score for audio data associated with the premises and indicating sound types, wherein the score for the audio data is determined based on:
a quantity of common sound types, wherein the common sound types are sound types that are indicated by the audio data and by the one of the plurality of audio signatures,
an order, indicated by the audio data, of the common sound types, and
an order, indicated by the one of the plurality of audio signatures, of the common sound types; and
causing, based on the score for the one of the plurality of audio signatures and the score for the audio data, a device associated with the premises to perform an action.
16. The method of claim 15 , wherein the causing the device to perform the action comprises one or more of:
disarming a security system associated with the premises, or
causing a device associated with the premises to output content of a service associated with the one of the plurality of audio signatures.
17. The method of claim 15 , wherein the sound types indicated by the one of the plurality of audio signatures comprise one or more of: a human voice sound type, an animal sounds sound type, an inaudible mechanical sounds sound type, an audible mechanical sounds sound type, an alert sounds sound type, or an absence of sound sound type.
18. The method of claim 15 , wherein the one of the plurality of audio signatures indicates that multiple sound types correspond to at least one of multiple time intervals during a signature period.
19. A method comprising:
determining, based on first audio data received from a premises, a potential occurrence of an alarm-triggering event;
determining, based on the potential occurrence of the alarm-triggering event and based on a first sound type associated with the first audio data, to defer triggering an alarm;
during an alarm deferral period associated with the determining to defer triggering the alarm, receiving second audio data from the premises;
determining, based on a second sound type associated with the second audio data and indicative of an authorized entry of the premises, and based on comparing the second audio data with an audio signature associated with a sequence of actions performed by a user for each of a plurality of entries into the premises, whether the alarm-triggering event occurred; and
controlling, based on determining whether the alarm-triggering event occurred, an alarm system of premises.
20. The method of claim 19 , wherein the second audio data comprises a series of sound type indicators, and wherein the sound type indicators indicate types of sounds occurring during the alarm deferral period.
21. The method of claim 19 , wherein the controlling comprises one of:
triggering, based on a determination that the alarm-triggering event occurred, an alarm associated with the premises; or
ignoring, based on a determination that the alarm triggering event did not occur, the potential occurrence of the alarm-triggering event.
22. The method of claim 19 , wherein the sequence of actions performed by the user comprises at least one of turning a key, opening a door, closing a door, dropping a key, walking, looking at a phone, disarming a security system, opening a closet, closing a closet, interacting with a pet, listening to music, taking food from refrigerator, cooking, or washing food.
23. The method of claim 19 , wherein the determining to defer triggering the alarm is further based on determining that the first sound type associated with the first audio data is a glass broken sound.
24. The method of claim 19 , wherein the determining whether the alarm-triggering event occurred comprises determining, based on the second audio data matching the audio signature associated with the behavior pattern of the user enteringsequence of actions performed by the user for each of the plurality of entries into the premises, that the alarm-triggering event did not occur.