Abstract: The present invention improves the accuracy of language prediction. A history speech meta-information understanding unit 11 obtains a history speech meta-information vector from a word string of a preceding speech using a meta-information understanding device. A history speech embedding unit 12 converts the word string of the preceding speech and a speaker label into a history speech embedding vector. A speech unit combination vector construction unit 13 obtains a speech unit combination vector by combining the history speech meta-information vector and the history speech embedding vector. A speech sequence embedding vector calculation unit 14 converts a plurality of speech unit combination vectors obtained for the past speech sequences to a speech sequence embedding vector. A language model score calculation unit 15 calculates a language model score of a current speech from a word string of the current speech, a speaker label, and a speech sequence embedding vector.

Abstract: A keyword is extracted robustly despite a voice recognition result including an error. A model storage unit 10 stores a keyword extraction model that accepts word vector representations of a plurality of words as an input and extracts and outputs a word vector representation of a word to be extracted as a keyword. A speech detection unit 11 detects a speech part from a voice signal. A voice recognition unit 12 executes voice recognition on the speech part of the voice signal and outputs a confusion network which is a voice recognition result. A word vector representation generating unit 13 generates a word vector representation including reliability of voice recognition with regard to each candidate word for each confusion set. A keyword extraction unit 14 inputs the word vector representation of the candidate word to the keyword extraction model in descending order of the reliability and obtains the word vector representation of the keyword.

Abstract: A voice recognition device 10 includes: a phonological awareness feature amount extraction unit 11 that transforms an acoustic feature amount sequence of input voice into a phonological awareness feature amount sequence for the language 1 using a first model parameter group; a phonological awareness feature amount extraction unit 12 that transforms the acoustic feature amount sequence of the input voice into a phonological awareness feature amount sequence for the language 2 using a second model parameter group; a phonological recognition unit 13 that generates a posterior probability sequence from the acoustic feature amount sequence of the input voice, the phonological awareness feature amount sequence for the language 1, and the phonological awareness feature amount sequence for the language 2 using a third model parameter group; and a voice text transformation unit 14 that performs voice recognition based on the posterior probability sequence to output text of a voice recognition result.

Abstract: The vehicle notification apparatus comprises: an operation element operated by a driver to activate a turn signal indicator; an information obtaining device that obtains information on an object around the vehicle; an auditory notification device configured to give an auditory notification; a visual notification device configured to give a visual notification; and a control unit.

Abstract: The vehicle notification apparatus comprises: an operation element operated by a driver to activate a turn signal indicator; an information obtaining device that obtains information on an object around the vehicle; an auditory notification device configured to give an auditory notification; a visual notification device configured to give a visual notification; and a control unit.

Abstract: Learning data is generated automatically without manually applying rules. An acoustic model learning data generation device 20 includes a stochastic attribute label generation model 21 that generates attribute labels from a first model parameter group according to a first probability distribution; a stochastic phoneme sequence generation model 22 that generates a phoneme sequence from a second model parameter group and the attribute labels according to a second probability distribution; and a stochastic acoustic feature quantity sequence generation model 23 that generates an acoustic feature quantity sequence from a third model parameter group, the attribute labels, and the phoneme sequence according to a third probability distribution.

Abstract: A language model score calculation apparatus calculates a prediction probability of a word wi as a language model score of a language model based on a recurrent neural network. The language model score calculation apparatus includes a memory; and a processor configured to execute converting a word wi-1 that is observed immediately before the word wi into a word vector ?(wi-1); converting a speaker label ri-1 corresponding to the word wi-1 and a speaker label ri corresponding to the word wi into a speaker vector ?(ri-1) and a speaker vector ?(ri), respectively; calculating a word history vector si by using the word vector ?(wi-1), the speaker vector ?(ri-1), and a word history vector si-1 that is obtained when a prediction probability of the word wi-1 is calculated; and calculating a prediction probability of the word wi by using the word history vector si-1 and the speaker vector ?(ri).

Abstract: A learning apparatus comprises a learning part that learns an error correction model by a set of a speech recognition result candidate and a correct text of speech recognition for given audio data, wherein the speech recognition result candidate includes a speech recognition result candidate which is different from the correct text, and the error correction model is a model that receives a word sequence of the speech recognition result candidate as input and outputs an error correction score indicating likelihood of the word sequence of the speech recognition result candidate in consideration of a speech recognition error.

Abstract: A radio wave absorbing material including a fluoropolymer. The fluoropolymer contains a vinylidene fluoride unit, and the radio wave absorbing material absorbs a radio wave having a frequency in a range of 1 MHz to 100 MHz.

Abstract: By using training data containing tuples of texts for M types of tasks in N types of languages and correct labels of the texts as input, an optimized parameter group that defines N inter-task shared transformation functions ?(n) corresponding to the N types of languages n and M inter-language shared transformation functions ?(m) corresponding to the M types of tasks in is obtained. At least one of N and M is an integer greater than or equal to 2, each ?(n) outputs a latent vector, which corresponds to the contents of an input text in a certain language n but does not depend on the language n, to ?(1), . . . ?(M), and each ?(m) uses, as input, the latent vector output from any one of ?(1), . . . ?(N) and outputs an output label corresponding to the latent vector for a certain task in.

Abstract: A power conversion device includes: a three-phase inverter including a plurality of switching elements; and a controller configured to PWM-control the three-phase inverter based on three phase voltage commands. The controller generates a zero-phase voltage command using a two-phase modulation scheme and a third harmonic component of the three phase voltage commands. The controller adds the generated zero-phase voltage command to each of the three phase voltage commands to thereby correct each of the three phase voltage commands. The controller compares each of the corrected three phase voltage commands with a carrier wave, to thereby generate a control signal for controlling switching of the plurality of switching elements.

Abstract: A film having a dielectric dissipation factor at a frequency of 1 kHz and 160° C. of 0.02% or lower and a dielectric breakdown strength at 160° C. of 400 V/?m or higher. Also disclosed is a film including at least one fluoropolymer selected from a tetrafluoroethylene/perfluoro(alkyl vinyl ether) copolymer and a tetrafluoroethylene/hexafluoropropylene copolymer, the fluoropolymer having a crystallinity of 65% or higher.

Abstract: A composition includes cells of Lactobacillus paracasei MCC1849 (NITE BP-01633). The composition is in the form of a pharmaceutical composition, a food, a drink, or a feed. The pharmaceutical composition can be used for treating an inflammatory disease, ulcer, food allergies, bronchial asthma, urticaria, pollinosis, anaphylactic shock or an opportunistic infection.

Abstract: A storage device includes: a first disabling unit configured to output write enable signals without change when at least two write addresses of a plurality of write addresses for which write enable signals are enabled and held by a first holding unit do not match; a second holding unit configured to hold sets of the plurality of write addresses held by the first holding unit and the plurality of write enable signals output by the first disabling unit; a second disabling unit configured to output one write enable signal of the plurality of write enable signals held by the second holding unit without change in a test mode; and a third holding unit configured to write data in accordance with sets of the plurality of write addresses held by the second holding unit and the plurality of write addresses output by the second disabling unit.

Abstract: A method for prophylaxis or treatment of a disease includes administering an an effective amount of an agent that contains cells of Lactobacillus paracasei MCC1849 (NITE BP-01633) to a subject in need of such prophylaxis or treatment. The disease can be one or more of an inflammatory disease, ulcer, food allergies, bronchial asthma, urticaria, pollinosis, anaphylactic shock or an opportunistic infection.

Abstract: A generating apparatus according to an embodiment of the present application includes an obtaining unit and a generating unit. The obtaining unit obtains intermediate images in an intermediate layer of a neural network that recognizes the quantity of objects contained in an image. The generating unit generates a combination image by combining together the intermediate images obtained by the obtaining unit. For example, the generating unit generates the combination image by combining together the intermediate images to which a weight has been applied on the basis of a change in the recognition of the quantity of objects corresponding to a process applied to each of the intermediate images.

Abstract: A memory circuit includes, a latch circuit that includes a first node, a second node, a first inverter having an input coupled to the first node and an output coupled to the second node, and a second inverter having an input coupled to the second node and an output coupled to the first node, a writing circuit that includes a first transistor coupled to the first node and a second transistor coupled to the second node and executes writing to the latch circuit using the first transistor and the second transistor, a command circuit detects the execution of the writing to the latch circuit and output a command signal to increase the potential of the second node before the termination of the writing of a low level to the first node, and a potential control circuit increases the potential of the second node based on the command signal.

Abstract: A magnetic body simulation device includes a memory; and a processor coupled to the memory and the processor configured to decide a search section for searching a rotational coefficient of each magnetization vector in a plurality of elements included in a magnetic body, in a process of calculating the each rotational coefficient in a state in which magnetic energy of the magnetic body is minimized, to determine whether or not a first condition is satisfied in which a width of the decided search section is less than or equal to a certain length, and a first rotational coefficient decided by a predetermined method is included in the search section, and to calculate, when the first condition is satisfied, a static magnetic field vector corresponding to the first rotational coefficient by linear interpolation based on static magnetic field vectors at both ends of the search section.

Abstract: A storage device includes: a first disabling unit configured to output write enable signals without change when at least two write addresses of a plurality of write addresses for which write enable signals are enabled and held by a first holding unit do not match; a second holding unit configured to hold sets of the plurality of write addresses held by the first holding unit and the plurality of write enable signals output by the first disabling unit; a second disabling unit configured to output one write enable signal of the plurality of write enable signals held by the second holding unit without change in a test mode; and a third holding unit configured to write data in accordance with sets of the plurality of write addresses held by the second holding unit and the plurality of write addresses output by the second disabling unit.

Abstract: An information processing device includes: a processor: calculates, for each mesh of a magnetic body, a differential vector between a magnetization vector before a saddle point and a magnetization vector after the saddle point; calculates, as a tangential vector, a component obtained by projecting the differential vector onto a surface; calculates, as a gradient vector, a component obtained by performing projection on the surface with respect to a vector obtained by differentiating energy of the magnetic body; calculates a velocity vector; calculates a fluctuation amount of the magnetization vector, a magnetization vector after the time, and a maximum of the fluctuation amount; resets the magnetization vector after the time to the magnetization vector; calculates a fluctuation amount of the reset magnetization vector, a maximum of the fluctuation amount, and the reset magnetization vector; and outputs information concerning energy of the magnetic body and a magnetization vector at the saddle point.