Enhanced audiovisual analytics

1. A computer-implemented method, comprising:

receiving audio data from a sensor of a monitoring system;

determining, using the audio data, an audio feature using an audio analytics classifier, wherein: the audio analytics classifier is an escalation classifier, and determining the audio feature using the audio analytics classifier comprises;

detecting an increase in an average signal power; and

determining, using the escalation classifier, whether the increase in the average signal power of the audio data over a time period satisfies a threshold;

determining whether an alert condition is satisfied using the audio feature that indicates whether the detected increased average signal power of the audio data over the time period satisfies the threshold;

in response to determining that the alert condition is satisfied, activating a camera of the monitoring system to obtain video data;

receiving the video data from the camera;

determining, using the video data, whether a security event associated with the alert condition exists; and

in response to determining that the security event associated with the alert condition exists, performing a monitoring action.

2. The method of claim 1 ,

wherein the audio analytics classifier is a discrete decibel threshold classifier, and

determining the audio feature using the audio analytics classifier comprises determining, using the discrete decibel threshold classifier, whether a signal power of the audio data satisfies a decibel threshold.

3. The method of claim 1 ,

wherein the audio analytics classifier is an average threshold classifier, and

determining the audio feature using the audio analytics classifier comprises determining, using the average threshold classifier, whether an average signal power of the audio data over a time period satisfies a decibel threshold.

4. The method of claim 1 ,

wherein the audio analytics classifier is an audio pattern classifier, and

determining the audio feature using the audio analytics classifier comprises identifying, using the audio pattern classifier, an audio pattern represented by the audio data over a time period.

5. The method of claim 1 , wherein:

determining whether the alert condition is satisfied using the audio data comprises at least one of:

determining whether at least one of a discrete decibel threshold, an average threshold, or an escalation threshold have been satisfied; or

determining whether an audio pattern exists in the audio data; and

activating the camera of the monitoring system to obtain the video data is responsive to determining that:

at least one of the discrete decibel threshold, the average threshold, or the escalation threshold has been satisfied; or

the audio pattern exists in the audio data.

6. The method of claim 5 , wherein determining that the alert condition is satisfied comprises determining that the audio data includes data indicative of an unsafe condition.

7. The method of claim 1 , wherein determining, using the video data, whether the security event associated with the alert condition exists comprises:

obtaining, from the camera, an image of conditions associated with the audio data received from the sensor; and

determining, using a set of rules, whether a combination of the image and the audio data corresponds to the security event.

8. The method of claim 7 , wherein the set of rules is configured by a neural network that is trained using user preference data and user interaction data.

9. A system comprising:

one or more computing devices; and

one or more storage devices storing instructions that, when executed by the one or more computing devices, cause the one or more computing devices to perform operations comprising:

receiving audio data from a sensor of a monitoring system;

determining, using the audio data, an audio feature using an audio analytics classifier, wherein: the audio analytics classifier is an escalation classifier, and determining the audio feature using the audio analytics classifier comprises:

detecting an increase in an average signal power; and

determining, using the escalation classifier, whether the increase in the average signal power of the audio data over a time period satisfies a threshold;

determining whether an alert condition is satisfied using the audio feature that indicates whether the detected increased average signal power of the audio data over the time period satisfies the threshold;

in response to determining that the alert condition is satisfied, activating a camera of the monitoring system to obtain video data;

receiving the video data from the camera;

determining, using the video data, whether a security event associated with the alert condition exists; and

in response to determining that the security event associated with the alert condition exists, performing a monitoring action.

10. The system of claim 9 ,

wherein the audio analytics classifier is a discrete decibel threshold classifier, and

determining the audio feature using the audio analytics classifier comprises determining, using the discrete decibel threshold classifier, whether a signal power of the audio data satisfies a decibel threshold.

11. The system of claim 9 ,

wherein the audio analytics classifier is an average threshold classifier, and

determining the audio feature using the audio analytics classifier comprises determining, using the average threshold classifier, whether an average signal power of the audio data over a time period satisfies a decibel threshold.

12. The system of claim 9 ,

wherein the audio analytics classifier is an audio pattern classifier, and

determining the audio feature using the audio analytics classifier comprises identifying, using the audio pattern classifier, an audio pattern represented by the audio data over a time period.

13. The system of claim 9 , wherein:

determining whether the alert condition is satisfied using the audio data comprises at least one of:

determining whether at least one of a discrete decibel threshold, an average threshold, or an escalation threshold have been satisfied; or

determining whether an audio pattern exists in the audio data; and

activating the camera of the monitoring system to obtain the video data is responsive to determining that:

at least one of the discrete decibel threshold, the average threshold, or the escalation threshold has been satisfied; or

the audio pattern exists in the audio data.

14. The system of claim 13 , wherein determining that the alert condition is satisfied comprises determining that the audio data includes data indicative of an unsafe condition.

15. The system of claim 9 , wherein determining, using the video data, whether the security event associated with the alert condition exists comprises:

obtaining, from the camera, an image of conditions associated with the audio data received from the sensor; and

determining, using a set of rules, whether a combination of the image and the audio data corresponds to the security event.

16. The system of claim 15 , wherein the set of rules is configured by a neural network that is trained using user preference data and user interaction data.

17. At least one non-transitory computer-readable storage device storing instructions that, when executed by one or more computers, causes the one or more computers to perform operations comprising:

receiving audio data from a sensor of a monitoring system;

determining, using the audio data, an audio feature using an audio analytics classifier, wherein:

the audio analytics classifier is an escalation classifier, and

determining the audio feature using the audio analytics classifier comprises:

detecting an increase in an average signal power; and

determining, using the escalation classifier, whether the increase in the average signal power of the audio data over a time period satisfies a threshold;

determining whether an alert condition is satisfied using the audio feature that indicates whether the detected increased average signal power of the audio data over the time period satisfies the threshold;

in response to determining that the alert condition is satisfied, activating a camera of the monitoring system to obtain video data;

receiving the video data from the camera;

determining, using the video data, whether a security event associated with the alert condition exists; and

in response to determining that the security event associated with the alert condition exists, performing a monitoring action.

18. The non-transitory computer-readable storage device of claim 17 ,

wherein the audio analytics classifier is a discrete decibel threshold classifier, and

determining the audio feature using the audio analytics classifier comprises determining, using the discrete decibel threshold classifier, whether a signal power of the audio data satisfies a decibel threshold.

19. The non-transitory computer-readable storage device of claim 17 , wherein:

determining whether the alert condition is satisfied using the audio data comprises at least one of:

determining whether at least one of a discrete decibel threshold, an average threshold, or an escalation threshold have been satisfied; or

determining whether an audio pattern exists in the audio data; and

activating the camera of the monitoring system to obtain the video data is responsive to determining that:

at least one of the discrete decibel threshold, the average threshold, or the escalation threshold has been satisfied; or

the audio pattern exists in the audio data.

20. The non-transitory computer-readable storage device of claim 19 , wherein determining that the alert condition is satisfied comprises determining that the audio data includes data indicative of an unsafe condition.