To provide an anomalous sound detection training technique by which a feature amount extraction function for detecting anomalous sound can be generated irrespective of whether training data for anomalous signals is available or not. An anomalous sound detection training apparatus includes: a first function updating unit 3 that updates a feature amount extraction function and an feature amount inverse transformation function, which are input, based on an optimization index of a variational autoencoder; an acoustic feature extraction unit 4 that extracts an acoustic feature of normal sound based on training data for normal sound; a normal sound model updating unit 5 that updates a normal sound model by using the acoustic feature that is extracted; a threshold updating unit 6 that obtains a threshold φ.sub.ρ corresponding to a false positive rate ρ, which has a predetermined value, by using the training data for normal sound and the feature amount extraction function that is input; and a second function updating unit 8 that updates the feature amount extraction function that is updated, based on a Neyman-Pearson-type optimization index defined by the threshold φ.sub.ρ that is obtained, and repeatedly performs processing of each of the above-mentioned units.

To provide an anomalous sound detection training technique by which a feature amount extraction function for detecting anomalous sound can be generated irrespective of whether training data for anomalous signals is available or not. An anomalous sound detection training apparatus includes: a first function updating unit 3 that updates a feature amount extraction function and an feature amount inverse transformation function, which are input, based on an optimization index of a variational autoencoder; an acoustic feature extraction unit 4 that extracts an acoustic feature of normal sound based on training data for normal sound; a normal sound model updating unit 5 that updates a normal sound model by using the acoustic feature that is extracted; a threshold updating unit 6 that obtains a threshold φ.sub.ρ corresponding to a false positive rate ρ, which has a predetermined value, by using the training data for normal sound and the feature amount extraction function that is input; and a second function updating unit 8 that updates the feature amount extraction function that is updated, based on a Neyman-Pearson-type optimization index defined by the threshold φ.sub.ρ that is obtained, and repeatedly performs processing of each of the above-mentioned units.

A damper is provided which can more reliably prevent malfunction and breakdown and which enables efficiently performing repair and inspection operations. This damper, provided with a casing linked to a first object and a rotating part linked to a second object rotatably attached to the first object, damps rotation in either the direction closing or the direction opening the second object, and is provided with a sensor which detects prescribed change in the external environment in the damper or around the damper, and a control unit which externally communicates, over a communication network, information relating to the change in the external environment detected by the sensor, wherein the sensor is configured from at least one of: a rotation sensor for detecting the number of revolutions of the rotating part: a sound sensor for detecting sound during rotations of the rotating part; a temperature sensor for detecting temperature; and a torque sensor for detecting torque on the basis of friction during rotation of the rotating part.

A damper is provided which can more reliably prevent malfunction and breakdown and which enables efficiently performing repair and inspection operations. This damper, provided with a casing linked to a first object and a rotating part linked to a second object rotatably attached to the first object, damps rotation in either the direction closing or the direction opening the second object, and is provided with a sensor which detects prescribed change in the external environment in the damper or around the damper, and a control unit which externally communicates, over a communication network, information relating to the change in the external environment detected by the sensor, wherein the sensor is configured from at least one of: a rotation sensor for detecting the number of revolutions of the rotating part: a sound sensor for detecting sound during rotations of the rotating part; a temperature sensor for detecting temperature; and a torque sensor for detecting torque on the basis of friction during rotation of the rotating part.

According to one embodiment, a learning apparatus includes a memory and a hardware processor connected to the memory which learns a transformation function to extract a feature value of an input signal. The hardware processor updates the transformation function based on a signal indicative of a first condition and a signal indicative of a second condition which is different from the first condition, using a first loss function on the signal indicative of the first condition and a second loss function on the signal indicative of the second condition. The second loss function is designed such that the second condition becomes distant from the first condition.

According to one embodiment, a learning apparatus includes a memory and a hardware processor connected to the memory which learns a transformation function to extract a feature value of an input signal. The hardware processor updates the transformation function based on a signal indicative of a first condition and a signal indicative of a second condition which is different from the first condition, using a first loss function on the signal indicative of the first condition and a second loss function on the signal indicative of the second condition. The second loss function is designed such that the second condition becomes distant from the first condition.

To determine, with high accuracy, whether an anomaly has occurred in equipment to be monitored, an anomaly detection apparatus includes an acquisition unit and a determination unit. The determination unit acquires detection result information indicating a detection result of a vibration sensor attached to equipment including a movable unit. The determination unit determines, by using the detection result information, presence or absence of an anomaly of the equipment. The detection result information includes a magnitude of a vibration of each of a plurality of axes oriented in directions different from each other. The determination unit computes, with respect to each of the plurality of axes, a difference between a magnitude of a vibration at a target timing, and a magnitude of a vibration at a predetermined period of time before the target timing or a criterion value previously set. Then, the determination unit determines, when the difference exceeds a criterion in any of the axes, that an anomaly is occurring in the equipment at the target timing.

To determine, with high accuracy, whether an anomaly has occurred in equipment to be monitored, an anomaly detection apparatus includes an acquisition unit and a determination unit. The determination unit acquires detection result information indicating a detection result of a vibration sensor attached to equipment including a movable unit. The determination unit determines, by using the detection result information, presence or absence of an anomaly of the equipment. The detection result information includes a magnitude of a vibration of each of a plurality of axes oriented in directions different from each other. The determination unit computes, with respect to each of the plurality of axes, a difference between a magnitude of a vibration at a target timing, and a magnitude of a vibration at a predetermined period of time before the target timing or a criterion value previously set. Then, the determination unit determines, when the difference exceeds a criterion in any of the axes, that an anomaly is occurring in the equipment at the target timing.

An anomalous sound detection training apparatus includes: a first acoustic feature extraction unit that extracts an acoustic feature of normal sound based on training data for normal sound by using an acoustic feature extractor; a normal sound model updating unit that updates a normal sound model by using the acoustic feature extracted; a second acoustic feature extraction unit that extracts an acoustic feature of anomalous sound based on simulated anomalous sound and extracts the acoustic feature of normal sound based on the training data for normal sound by using the acoustic feature extractor; and an acoustic feature extractor updating unit that updates the acoustic feature extractor by using the acoustic feature of anomalous sound and the acoustic feature of normal sound that have been extracted, in which processing by the units is repeatedly performed.

An anomalous sound detection training apparatus includes: a first acoustic feature extraction unit that extracts an acoustic feature of normal sound based on training data for normal sound by using an acoustic feature extractor; a normal sound model updating unit that updates a normal sound model by using the acoustic feature extracted; a second acoustic feature extraction unit that extracts an acoustic feature of anomalous sound based on simulated anomalous sound and extracts the acoustic feature of normal sound based on the training data for normal sound by using the acoustic feature extractor; and an acoustic feature extractor updating unit that updates the acoustic feature extractor by using the acoustic feature of anomalous sound and the acoustic feature of normal sound that have been extracted, in which processing by the units is repeatedly performed.