SYSTEM FOR DETERMINING AN EMOTION OF A USER

    公开(公告)号:US20220346682A1

    公开(公告)日:2022-11-03

    申请号:US17753920

    申请日:2020-09-21

    申请人: OVOMIND K.K.

    发明人: Yann Frachi

    IPC分类号: A61B5/16 A61B5/024 A61B5/00

    摘要: A method for calculating a numerical data pair representing an emotional state comprises: acquiring: ⋅a first series of physiological signals by a GSR and/or EDA electrodermal sensor; ⋅a second series of PPG physiological signals by a heart rate sensor; ⋅transmitting to a remote server the timestamped signals as well as an identifier of the acquisition device; and ⋅processing the signals to characterize an Sarousal, Svalence data pair. The processing of the first series of signals is of the empirical mode decomposition (EMD) type over a sliding time window, the result of which provides the first Sarousal value of the pair (arousal level). Processing of the second series of signals comprises a step of band-pass filtering of frequencies between 0.04 and 0.26 Hz and of peak detection and RR inter-peak time measurement, over the sliding time window, the result of which provides the second Svalence value of the pair (valence).

    METHOD FOR AUTOMATICALLY PREDICTING THE EMOTIONAL EFFECT PRODUCTED BY A VIDEO GAME SEQUENCE

    公开(公告)号:US20240245986A1

    公开(公告)日:2024-07-25

    申请号:US18564176

    申请日:2022-05-25

    申请人: OVOMIND K.K

    发明人: Yann FRACHI

    IPC分类号: A63F13/212 A63F13/67

    CPC分类号: A63F13/212 A63F13/67

    摘要: A method is provided for automatically predicting the emotional effect produced by a video game sequence, comprising labeling sequences of the game by automatically generating descriptors at time sequences of the game, the labeling comprising applying digital processing to the audio stream of the video game sequence using a neural network architecture and an NLP coding layer, to extract a first series of timestamped descriptors, and applying digital processing to the video stream to provide a second series of timestamped descriptors for characterizing the scenes of each image of the video stream, and transmitting them as M-tuples to a neural network. The method also comprises processing biosignals to extract timestamped signals and transmit them as N-tuples to a neural network and processing the M-tuples corresponding to the timestamped descriptors and the N-tuples to provide at least one indicator predicting the emotional state brought about by a type of audiovisual sequence.

    ELECTRODERMAL APPARATUS
    3.
    发明公开

    公开(公告)号:US20230371837A1

    公开(公告)日:2023-11-23

    申请号:US18027747

    申请日:2021-09-23

    申请人: OVOMIND K.K.

    发明人: Yann FRACHI

    摘要: The invention presents an electrodermal apparatus comprising at least two electrodes (10), each of which is configured to be in contact with a dermal region, and an electronic circuit comprising an analog-to-digital converter and a processor that is configured to execute computer-executable instructions in order to bias the electrodes (10), to measure the current flowing between pairs of electrodes (10), to apply digital preprocessing to the measured electrical signals and to determine the conductance of the dermal layer, wherein the sensor further comprises an accelerometer, delivering electrical signals to the processor, the preprocessing taking into account the level of the signals that are delivered by the accelerometer.