US 12,281,932 B2
Abnormal sound identification device, abnormal sound identification method, and non-transitory storage medium
Yu Ueda, Toyota (JP)
Assigned to TOYOTA JIDOSHA KABUSHIKI KAISHA, Toyota (JP)
Filed by TOYOTA JIDOSHA KABUSHIKI KAISHA, Toyota (JP)
Filed on Jul. 8, 2022, as Appl. No. 17/860,321.
Claims priority of application No. 2021-115923 (JP), filed on Jul. 13, 2021.
Prior Publication US 2023/0012559 A1, Jan. 19, 2023
Int. Cl. G01H 3/08 (2006.01); G01H 3/06 (2006.01); G06N 3/08 (2023.01)
CPC G01H 3/08 (2013.01) [G01H 3/06 (2013.01); G06N 3/08 (2013.01)] 5 Claims
OG exemplary drawing
 
1. An abnormal sound identification device comprising:
an arithmetic device configured to access a trained model of artificial intelligence; and
an output device,
wherein the arithmetic device is configured to
identify frequency-time data that indicates a temporal change of a frequency spectrum of a sound recorded in a vehicle,
specify a first time range and a second time range different from the first time range in the frequency-time data,
input the frequency-time data to the trained model to cause the trained model to identify one or more abnormal sounds generated in the first time range as a first abnormal sound based on the input frequency-time data and cause the trained model to identify one or more abnormal sounds generated in the second time range as a second abnormal sound based on the input frequency-time data, and
cause the output device to output a kind of the first abnormal sound with the kind not matching a kind of the second abnormal sound among the first abnormal sounds.
 
5. An abnormal sound identification device comprising:
an arithmetic device configured to access a trained model of artificial intelligence;
a storage device configured to store corresponding data in which a kind of an abnormal sound is associated with a component of a vehicle; and
an output device,
wherein the arithmetic device is configured to
identify frequency-time data indicating a temporal change in a frequency spectrum of a sound recorded in the vehicle and component operating state data indicating a temporal change of an operating state of the component of the vehicle during recording of the sound,
input the frequency-time data to the trained model to cause the trained model to identify an abnormal sound included in the sound as an identified abnormal sound based on the input frequency-time data, and
identify a corresponding component that is the component associated with the identified abnormal sound in the corresponding data, identify an operating time zone of the corresponding component based on the component operating state data of the corresponding component, and determine whether to cause the output device to output a kind of the identified abnormal sound in determination processing including determining whether a generation time zone of the identified abnormal sound and the operating time zone of the corresponding component overlap as at least one of determination elements.