System and method for ambient noise detection, identification and management
OwnerElina Birmingham, Siamak Arzanpour +1inventors
InventorsElina Birmingham, Siamak Arzanpour, Behnaz Bahmei
Life of this patent
Estimating the term…
Abstract
Examples of system for ambient aversive sound detection, identification and management are described. The system comprises an earpiece device with a microphone configured to capture ambient sound around a user and sample it into small segments of the ambient sound, a speaker and a regulator to regulate the ambient sound segment transmitted to the speaker. The system further comprises a processing unit that identifies aversive ambient sound signals in the captured sound segment and provide recommendation action to manage the aversive sound signal by removing, supressing, attenuating or masking the aversive signals.
Claims28
System for ambient aversive sound detection, identification and management, the system comprising:
- an earpiece device comprising a microphone configured to capture ambient sound around a user, the microphone is configured to capture samples of small segments of the ambient sound, and a speaker;
- an interface having a hardware processor programmed with executable instructions for obtaining input data, transmitting data and providing output data;
- a memory unit storing input information, a library of aversive ambient sound signals, aversive sound feature maps, identifying prediction models, aversive sound identifying classes, and aversive sound suppression prediction models;
- and
- a processing unit in communication with the earpiece device, the memory unit and the interface and comprising:
- an identifying unit coupled to the earpiece device and the memory unit and programmed with executable instructions to identify an aversive ambient sound signal in the ambient sound segment by extracting at least one feature of the sound in the sound segment, classify such ambient sound segment as a mix-signal of ambient sound segments with aversive sound signal and create a feature map of such sound segment, the feature map of the sound segment is processed using the identifying prediction model stored in the memory unit to compare at least one feature in the feature map with the feature maps of aversive sound signals in the memory unit, wherein when the aversive ambient sound is identified the identifying unit categorizes the aversive sound signal with an identifying class;
- and
- a filtration processing unit coupled to the identifying unit and the memory unit and programmed with executable instructions to receive the mix-signal of ambient sound segments with aversive sound signal and process the mix-signal computing an amplitude and phase of the mix-signal to generate a feature map of the mix-signal, compare the feature map with the stored feature maps using at least one aversive sound suppressive model and provide a signal of a recommend action through the earpiece device and/or the interface to manage such aversive sound signals.
A method for ambient aversive sound detection, identification and management, the method comprising:
- capturing an ambient sound around a user using a microphone in an earpiece device, the microphone being used to capture samples of small segments of the ambient sound;
- storing input information, a library of aversive ambient sound signals, aversive sound feature maps, identifying prediction models, aversive sound identifying classes and aversive sound suppression prediction models on a memory unit;
- and
- processing the captured sound segments by a processing unit, the processing step comprises:
- extracting at least one feature of the sound signal in the sound segment, creating a feature map of the sound segment, comparing the at least one feature in the feature map with the feature maps of aversive sound signals in the identifying prediction model stored in the memory unit, identifying an aversive sound signal in the captured sound segment using the identifying prediction model and categorizing the identified aversive sound signal with an identifying class;
- and
- filtration processing of a mix-signal that comprises the ambient sound segments with the aversive ambient sound signal, computing an amplitude and phase of the mix-signal to generate a feature map of the mix-signal, comparing the feature map with the stored feature maps using at least one aversive sound suppressive model and providing a recommend action to manage the aversive sound signals.
Dependencies are inferred from the claim text.