Listening devices for obtaining metrics from ambient noise
AssigneeCapital One Services, LLC
InventorsMichael Mossoba, Joshua Edwards, Abdelkadar M'Hamed Benkreira, Austen Novis, Sophie Bermudez
Life of this patent
Estimating the term…
Abstract
A device may receive audio data based on a capturing of sounds associated with a structure. The device may obtain a model associated with the structure. The model may have been trained to receive the audio data as input, determine a score that identifies a likelihood that a sound is present in the audio data, and identify the sound based on the score. The device may determine at least one parameter associated with the sound. The device may generate a metric based on the at least one parameter associated with the sound, and perform an action based on the metric.
Claims20
A method, comprising:
- receiving, by a device, audio data based on sounds associated with a structure;
- obtaining, by the device, a model associated with the structure,
- the model having been trained to:
- receive the audio data as input, and
- identify a sound as output;
- identifying, by the device and based on the model, the sound;
- determining, by the device, at least one parameter associated with the sound;
- generating, by the device, a first metric based on the at least one parameter,
- the first metric indicating usage of an equipment associated with the structure;
- obtaining, by the device, at least one second metric associated with the equipment,
- the at least one second metric indicating usage of other equipment associated with at least one other structure;
- and
- performing, by the device, an action based on the first metric and the at least one second metric.
A device comprising:
- one or more memories;
- and
- one or more processors, communicatively coupled to the one or more memories, configured to:
- receive audio data based on sounds associated with a structure;
- identify a particular sound included in the audio data;
- identify, based on the particular sound, an equipment associated with the structure;
- determine, based on the particular sound and the equipment, a parameter associated with the equipment,
- the parameter indicating that the equipment is in use;
- generate a first metric based on the parameter,
- the first metric indicating at least one of:
- a period of time the equipment is in use, or
- a quantity of uses of the equipment;
- obtain, from a second device, a second metric associated with the equipment,
- the second metric indicating usage of other equipment associated with at least one other structure;
- and
- perform an action based on the first metric and the second metric.
A non-transitory computer-readable medium storing instructions, the instructions comprising:
- one or more instructions that, when executed by one or more processors, cause the one or more processors to:
- receive audio data associated with a structure;
- identify, based on the audio data and a model, an equipment associated with the structure,
- the model having been trained to:
- receive the audio data as input, and
- identify the equipment as output;
- determine, based on the audio data and the equipment, a parameter associated with the equipment,
- the parameter indicating that the equipment is in use;
- generate a first metric based on the parameter,
- the first metric indicating at least one of:
- a period of time the equipment is in use, or
- a quantity of uses of the equipment;
- obtain a second metric associated with the equipment,
- the second metric indicating usage of other equipment associated with at least one other structure;
- and
- perform an action based on the first metric and the second metric.
Dependencies are inferred from the claim text.