Embedded audio sensor system and methods
An embedded sensor can include an audio detector, a digital signal processor, a library, and a rules engine. The digital signal processor can be configured to receive signals from the audio detector and to identify the environment in which the embedded sensor is located. The library can store statistical models associated with specific environments, and the digital signal processor can be configured identify specific events based on detected sounds within the particular environment by utilizing the statistical model associated with the particular environment. The DSP can associate a probability of accuracy for the identified audible event. A rules engine can be configured to receive the probability and transmit a report of the detected audible event.
1. An embedded sensor comprising:
an audio detector;
a processor configured to receive signals from the audio detector;
a library for storing an ensemble of statistical models associated with an environment, wherein the processor is configured to analyze a cost function to determine an optimal statistical model from the ensemble, wherein the processor is configured to associate a probability with a detected audible event based on the optimal statistical model; and
a rules engine configured to receive the probability associated with the detected audible event and identify the detected audible event.
2. The embedded sensor of claim 1 , further comprising a second detector configured to send detector signals to the processor, wherein the processor is configured to base the probability on the detected audible event and on the detector signals.
3. The embedded sensor of claim 2 , wherein the second detector comprises a camera.
4. The embedded sensor of claim 2 , wherein the second detector comprises a thermometer and a humidity detector.
5. The embedded sensor of claim 1 , further comprises a second detector configured to send data to the rules engine, wherein the rules engine is configured to infer an event based on the probability and on the data.
6. An embedded sensor for identifying sounds comprising:
a remote sensor having an audio detector, a digital signal processor, and a library, wherein the remote sensor is remotely connected to a cloud;
the digital signal processor is communicatively coupled to the audio detector;
wherein the remote sensor is configured to classify an environment and to determine the audible event based on a statistical model associated with the environment; and
wherein the digital signal processor is configured to analyze a cost function associated with the statistical model to determine an optimal statistical model.
7. The embedded sensor of claim 6 , wherein the remote sensor is configured to receive updates from the cloud through a service.
8. The embedded sensor of claim 7 , wherein the updates include a new statistical model.
9. The embedded sensor of claim 7 , wherein the updates include firmware.
10. The embedded sensor of claim 6 , wherein the probability is based on the optimal statistical model.
11. The embedded sensor of claim 10 , further comprising a rules engine configured to receive the probability and to determine whether transmit a report of the audible event.
12. The embedded sensor of claim 11 , wherein the rules engine is configured to transmit the report based on a signal-to-noise ratio.