Abstract: In order to enable a user to efficiently label a region of a target object in an image, a labeling device (100) includes: a display processing section (11) which causes a first target image and a color palette to be displayed; and a marking processing section (13) which marks a region on the first target image and a region on the color palette in response to an input from the user. In response to the input from the user with respect to one of the region on the first target image and the region on the color palette, the marking processing section (13) marks the other.

Abstract: The state estimation device includes a generation unit and an estimation unit. The generation unit applies an acoustic stimulus to an object and generates an impulse response of the object. The estimation unit estimates the state of the object by deep learning using the learned model, based on the impulse response.

Abstract: A frequency setting unit (120) acquires information representing required sound quality for sound output from parametric speakers (140). The frequency setting unit (120) performs determination about whether or not to make the frequency of a modulation signal mutually different among a plurality of parametric speakers (140) based on information representing the acquired sound quality. The frequency setting unit (120) sets the frequency of the modulation signal in each of a plurality of parametric speakers (140) based on the determination result. The modulation unit (130) generates modulation signals having the frequencies set by the frequency setting unit (120). The modulation signals are input to the plurality of parametric speakers (140).