Abstract: A model is learned that is capable of accurate label estimation even if learning data is used for which the number of evaluators per piece of data is small.

Abstract: Paralinguistic information is estimated with high accuracy even when an utterance for which it is difficult to identify paralinguistic information is used for model learning. An acoustic feature extraction unit 11 extracts an acoustic feature from an utterance. An anti-teacher decision unit 12 decides, based on a paralinguistic information label indicating a determination result of paralinguistic information given by a plurality of listeners for each utterance, an anti-teacher label indicating an anti-teacher serving as incorrect paralinguistic information for the utterance. An anti-teacher estimation model learning unit 13 learns, based on an acoustic feature extracted from the utterance and the anti-teacher label, an anti-teacher estimation model for outputting a posterior probability of anti-teacher for an input acoustic feature.

Abstract: A model capable of estimating a label with high accuracy is learned even when training data involving a small number of raters per data item is used. Learning processing is performed in which a plurality of data items and label expectation values that are indicators representing degrees of correctness of individual labels on the data items are used in pairs as training data, and a model that estimates a label on an input data item is obtained.

Abstract: An electron beam inspection apparatus includes a plurality of electrodes to surround an inspection substrate placed on a stage, a camera to measure, for each of the plurality of electrodes, a gap between a peripheral edge of the inspection substrate and an electrode of the plurality of electrodes, a retarding potential application circuit to apply a retarding potential to the inspection substrate, an electrode potential application circuit to apply, to each electrode, a corresponding potential of potentials each obtained by adding an offset potential, which is variable according to a measured gap, to the retarding potential to be applied to the inspection substrate, and an electron optical system to irradiate the inspection substrate with electron beams, in the state where the retarding potential has been applied to the inspection substrate and the corresponding potential of the potentials has been individually applied to each of the plurality of electrodes.

Abstract: To increase the accuracy of paralinguistic information estimation. A paralinguistic information estimation model storage unit 20 stores a paralinguistic information estimation model outputting, with a plurality of independent features as inputs, paralinguistic information estimation results. A feature extraction unit 11 extracts the features from an input utterance. A paralinguistic information estimation unit 20 estimates paralinguistic information of the input utterance from the features extracted from the input utterance, by using the paralinguistic information estimation model.

Abstract: An attribute identification technology that can reject an attribute identification result if the reliability thereof is low is provided. An attribute identification device includes: a posteriori probability calculation unit 110 that calculates, from input speech, a posteriori probability sequence {q(c, i)} which is a sequence of the posteriori probabilities q(c, i) that a frame i of the input speech is a class c; a reliability calculation unit 120 that calculates, from the posteriori probability sequence {q(c, i)}, reliability r(c) indicating the extent to which the class c is a correct attribute identification result; and an attribute identification result generating unit 130 that generates an attribute identification result L of the input speech from the posteriori probability sequence {q(c, i)} and the reliability r(c).

Abstract: A paper sheet storage apparatus (1) includes: a placement mount (10) on which bills (B) , i.e., an example of paper sheets, are placed; a leading-edge position restriction guide (20) that restricts the position of the leading edge portion of a bill (B) in an entering direction (D) in which the bill (B) advances when being transported onto the placement mount (10); and a sensor (30) that detects the presence/absence of an accumulation space (S) for bills (B) that is located above the placement mount (10). The sensor (30) detects the presence/absence of an accumulation space (S) by using detection light (L) emitted to a space between a first position (P1) that is located on an opposite side of the leading-edge position restriction guide (20) from the bills (B) on the placement mount (10) and a second position (P2) below the first position (P1) that is located on an opposite side of the bills (B) on the placement mount (10) from the leading-edge position restriction guide (20).

Abstract: A multiple electron beam irradiation apparatus includes a shaping aperture array substrate to form multiple primary electron beams, a plurality of electrode array substrates stacked each to dispose thereon a plurality of electrodes each arranged at a passage position of each of the multiple primary electron beams, each of the multiple primary electron beams surrounded by an electrode of the plurality of electrodes when each of the multiple primary electron beams passes through the passage position, the first wiring and the second wiring applied with one of different electric potentials, and a stage to mount thereon a target object to be irradiated with the multiple primary electron beams having passed through the plurality of electrode array substrates, wherein, in each of the plurality of electrode array substrates, each of the plurality of electrodes is electrically connected to either one of the first wiring and the second wiring.

Abstract: A multiple electron beam irradiation apparatus includes a forming mechanism which forms multiple primary electron beams; a plurality of electrode substrates being stacked in each of which a plurality of openings of various diameter dimensions are formed, the plurality of openings being arranged at passage positions of the multiple primary electron beams, and through each of which a corresponding one of the multiple primary electron beams passes, the plurality of electrode substrates being able to adjust an image plane conjugate position of each of the multiple primary electron beams depending on a corresponding one of the various diameter dimensions; and a stage which is capable of mounting thereon a target object to be irradiated with the multiple primary electron beams having passed through the plurality of electrode substrates.

Abstract: An attribute identification technology that can reject an attribute identification result if the reliability thereof is low is provided. An attribute identification device includes: a posteriori probability calculation unit 110 that calculates, from input speech, a posteriori probability sequence {q(c, i)} which is a sequence of the posteriori probabilities q(c, i) that a frame i of the input speech is a class c; a reliability calculation unit 120 that calculates, from the posteriori probability sequence {q(c, i)}, reliability r(c) indicating the extent to which the class c is a correct attribute identification result; and an attribute identification result generating unit 130 that generates an attribute identification result L of the input speech from the posteriori probability sequence {q(c, i)} and the reliability r(c).

Abstract: An estimation model is self-trained by utilizing a large amount of utterance with no teacher label. An estimation model learning part (11) learns an estimation model for estimating confidence for each of predetermined labels from each of feature amounts extracted from input data using a plurality of the independent feature amounts extracted from utterance with a teacher label. A paralinguistic information estimating part (12) estimates confidence for each of the labels from feature amounts extracted from utterance with no teacher label using the estimation model. When confidence for each label obtained from the utterance with no teacher label exceeds all confidence thresholds which are set in advance for each of the feature amounts for the feature amount to be learned, a data selecting part (13) adds a label corresponding to the confidence to data with no teacher label as a teacher label to select the data as self-training data.

Abstract: An electron beam inspection apparatus includes a plurality of electrodes to surround an inspection substrate placed on a stage, a camera to measure, for each of the plurality of electrodes, a gap between a peripheral edge of the inspection substrate and an electrode of the plurality of electrodes, a retarding potential application circuit to apply a retarding potential to the inspection substrate, an electrode potential application circuit to apply, to each electrode, a corresponding potential of potentials each obtained by adding an offset potential, which is variable according to a measured gap, to the retarding potential to be applied to the inspection substrate, and an electron optical system to irradiate the inspection substrate with electron beams, in the state where the retarding potential has been applied to the inspection substrate and the corresponding potential of the potentials has been individually applied to each of the plurality of electrodes.

Abstract: An attribute identification technology that can reject an attribute identification result if the reliability thereof is low is provided. An attribute identification device includes: a posteriori probability calculation unit 110 that calculates, from input speech, a posteriori probability sequence {q(c, i)} which is a sequence of the posteriori probabilities q(c, i) that a frame i of the input speech is a class c; a reliability calculation unit 120 that calculates, from the posteriori probability sequence {q(c, i)}, reliability r(c) indicating the extent to which the class c is a correct attribute identification result; and an attribute identification result generating unit 130 that generates an attribute identification result L of the input speech from the posteriori probability sequence {q(c, i)} and the reliability r(c).

Abstract: Provided is a multiple electron beam inspection apparatus including: an irradiation source irradiating a substrate with multiple electron beams; a stage on which is cable of mounting the substrate; an electromagnetic lens provided between the irradiation source and the stage, the electromagnetic lens generating a lens magnetic field, the multiple electron beams being capable of passing through the lens magnetic field; an electrostatic lens provided in the lens magnetic field, the electrostatic lens including a plurality of through-holes and a plurality of electrodes, the plurality of through-holes having wall surfaces respectively, each of the multiple electron beams being capable of passing through the corresponding each of the plurality of through-holes, each of the plurality of electrodes provided on each of the wall surfaces of the plurality of through-holes, at least one of the through-holes provided apart from a central axis of trajectory of the multiple electron beams having a spiral shape; and a power

Abstract: A paper sheet separation device includes a separation roller coming into contact with paper sheets, a feed roller rotating with respect to the separation roller to convey the paper sheets separately one by one, and a pair of guide members having guide faces coming into sliding contact with a front end of the paper sheets and being provided on both sides in an axial direction of the separation roller. A friction coefficient of the guide face is smaller than a friction coefficient of a peripheral face of the separation roller. The guide face is provided protruding outward of the separation roller from the peripheral face of the separation roller in a radial direction of the separation roller. The guide face is supported to be able to move inward in the radial direction of the separation roller at a contact point between the separation roller and the feed roller.

Abstract: A multiple beam inspection apparatus includes a multi-detector to detect multiple secondary electron beams generated because a target object is irradiated with multiple primary electron beams, and to include plural detection pixels each receiving irradiation of a corresponding one of the multiple secondary electron beams, and having a region which receives irradiation of a corresponding secondary electron beam and is larger than the irradiation spot size of the corresponding secondary electron beam, a shifting mechanism to shift irradiation positions of the multiple secondary electron beams irradiating the plural detection pixels, a determination circuitry to determine whether sensitivity of at least one of the plural detection pixels is degraded, and a setting circuitry to set, when sensitivity of at least one detection pixel is degraded, irradiation position shifting destinations of multiple secondary electron beams, irradiating the plural detection pixels, to be within respective corresponding same detecti

Abstract: An urgency level estimation technique of estimating an urgency level of a speaker for free uttered speech, which does not require a specific word, is provided. An urgency level estimation apparatus includes a feature amount extracting part configured to extract a feature amount of an utterance from uttered speech, and an urgency level estimating part configured to estimate an urgency level of a speaker of the uttered speech from the feature amount based on a relationship between a feature amount extracted from uttered speech and an urgency level of a speaker of the uttered speech, the relationship being determined in advance, and the feature amount includes at least one of a feature indicating speaking speed of the uttered speech, a feature indicating voice pitch of the uttered speech and a feature indicating a power level of the uttered speech.

Abstract: With correct emotion classes selected as correct values of an emotion of an utterer of a first utterance from among a plurality of emotion classes C1, . . . , CK by listeners who have listened to the first utterance, as an input, the numbers of times ni that emotion classes Ci have been selected as the correct emotion classes are obtained, and rates of the numbers of times nk to a sum total of the numbers of times n1, . . . , nK or smoothed values of the rates are obtained as correct emotion soft labels tk(s) corresponding to the first utterance.

Abstract: A multiple beam image acquisition apparatus includes a stage to mount thereon a target object, a beam forming mechanism to form multiple primary electron beams and a measurement primary electron beam, a primary electron optical system to collectively irradiate the target object surface with the multiple primary electron beams and the measurement primary electron beam, a secondary electron optical system to collectively guide multiple secondary electron beams generated because the target object is irradiated with the multiple primary electron beams, and a measurement secondary electron beam generated because the target object is irradiated with the measurement primary electron beam, a multi-detector to detect the multiple secondary electron beams collectively guided, a measurement mechanism to measure a position of the measurement secondary electron beam collectively guided, and a correction mechanism to correct a trajectory of the multiple secondary electron beams by using a measured position of the measureme