Abstract: A method and device for encoding/decoding an audio signal based on dequantization through potential diffusion are provided. The method of decoding an audio signal includes obtaining a discrete latent vector in which a speech signal is quantized and based on the discrete latent vector, outputting a continuous latent vector in which the discrete latent vector is dequantized.

Abstract: A method and apparatus for encoding/decoding a neural network-based personalized speech are provided. The method includes outputting a first bit stream in which an input speech signal is encrypted, based on the input speech signal, and outputting a second bit stream in which speaker information of the input speech signal is encrypted, based on the input speech signal.

Abstract: Disclosed are a device and method for audio signal processing. The audio signal processing device according to an embodiment includes a receiver configured to receive a bitstream corresponding to a compressed audio signal and a processor. The processor may be configured to generate a real restoration signal or a complex restoration signal by performing inverse quantization on real data of the bitstream or complex data of the bitstream, generate a result of real Frequency Domain Noise Shaping (FDNS) synthesis or a result of complex FDNS synthesis by performing FDNS synthesis on the real restoration signal or the complex restoration signal, and generate a restored audio signal by performing frequency-to-time transform on the result of the real FDNS synthesis or the result of the complex FDNS synthesis.

Abstract: An embodiment joint structure for a vehicle body includes a front body module that is a stationary part fixable to a front part of an under body of the vehicle body, a rear body module that is a variable part having a preset shape, the rear body module being attachable to and detachable from a rear part of the under body and a rear part of the front body module, a plurality of stationary part coupling modules disposed on an upper end surface portion of a rear edge of the front body module, and a plurality of variable part coupling modules disposed on an upper end surface portion of a front edge of the rear body module and couplable to the stationary part coupling modules by an electromagnetic force.

Abstract: Provided are a separator and an electrochemical device including the same. According to an embodiment of the present disclosure, a separator is provided which includes a porous substrate; and an inorganic particle layer which is placed on one or both surfaces of the porous substrate and includes inorganic particles and an inorganic binder, wherein the inorganic particle layer includes an organic compound having a lower content than the inorganic binder and the inorganic binder includes a metal silicate.

Abstract: An embodiment joint structure of a vehicle body includes a front body module fixed to a front portion of a center floor member of an underbody of the vehicle body, wherein the front body module is a fixed part, a rear body module detachably fixed to a rear portion of the center floor member and a rear portion of the front body module in a set shape, wherein the rear body module is a variable part, a plurality of center floor member coupling modules disposed on the center floor member at the rear portion of the front body module, and a plurality of variable part coupling modules disposed at a lower side of the rear body module and configured to be coupled to the center floor member coupling modules by electromagnetic force.

Abstract: An analog-to-digital converter is provided. An analog-to-digital converter comprises an interleaver receiving and processing an input signal that is an analog signal and a plurality of sub-ADCs, wherein the interleaver includes a reference circuit outputting a second voltage based on a first voltage, a high-pass filter receiving the second voltage and outputting a first signal obtained by changing a common mode voltage of the input signal to the second voltage, a sampling circuit generating a second signal obtained by sampling an alternate current component of the first signal, and a buffer outputting a third signal obtained by buffering the second signal by using a buffering circuit, and the sub-ADC converts the third signal into a digital signal by using the first voltage.

Abstract: A herd behavior pattern analysis apparatus of video-based herd objects includes a data transmission/reception module; a memory that stores a herd pattern analysis program of the video-based herd objects; and a processor that executes the program stored in the memory, in which the program performs video pre-processing to detect an edge image of the herd object based on an input video captured through at least one camera allocated to a space where the herd objects are accommodated, and inputs the edge image into a herd pattern analysis model to detect pattern information of the herd object and to determine whether the herd object is normal based on the pattern information, and the herd pattern analysis model is a model learned using learning data including the edge image of each herd object, and outputs pattern information of the herd object based on the input video.

Abstract: The present invention relates to a composition for promoting myogenesis, containing, as an active ingredient, a processed ginseng extract in which a trace amount of a ginsenoside ingredient is increased. It has been ascertained that the processed ginseng extract promotes the differentiation of myoblasts into muscle and inhibits muscle atrophy caused by myostatin, which is a myogenesis inhibitory factor, and thus it is expected that a composition for preventing or treating muscle disorder-related diseases, having excellent effects, can be developed.

Abstract: Artificial blood for reproducing an expectorated blood spatter according to an embodiment includes maple syrup, water, and food coloring.

Abstract: A method of mathematically estimating the center of a position trajectory of an object generating a swing cast-off spatter includes obtaining position coordinates of a plurality of bloodstains to be analyzed, calculating collision angles of the plurality of bloodstains, generating straight lines based on the position coordinates and collision angles of the plurality of bloodstains, generating a triangle by selecting three of the generated straight lines, and generating an escribed circle of the triangle and calculating center coordinates and a radius length of the escribed circle.

Abstract: A semiconductor system includes a controller configured to apply a command address, a first chip selection signal, and a second chip selection signal, and a semiconductor device including a first rank and a second rank configured to calibrate each termination resistance, based on the command address, the first chip selection signal, and the second chip selection signal.

Abstract: Provided is a method and apparatus for designing and testing an audio codec using quantization based on white noise modeling. A neural network-based audio encoder design method includes generating a quantized latent vector and a reconstructed signal corresponding to an input signal by using a white noise modeling-based quantization process, computing a total loss for training a neural network-based audio codec, based on the input signal, the reconstruction signal, and the quantized latent vector, training the neural network-based audio codec by using the total loss, and validating the trained neural network-based audio codec to select the best neural network-based audio codec.

Abstract: An encoding method, a decoding method, an encoder for performing the encoding method, and a decoder for performing the decoding method are provided. The encoding method includes outputting LP coefficients bitstream and a residual signal by performing an LP analysis on an input signal, outputting a first latent signal obtained by encoding a periodic component of the residual signal, a second latent signal obtained by encoding a non-periodic component of the residual signal, and a weight vector for each of the first latent signal and the second latent signal, using a first neural network module, and outputting a first bitstream obtained by quantizing the first latent signal, a second bitstream obtained by quantizing the second latent signal, and a weight bitstream obtained by quantizing the weight vector, using a quantization module.

Abstract: The present disclosure provides a location information acquisition method of a computing device, including the steps of: acquiring first location information at which an electro-optical topographic surveying device is located in response to a received satellite Global Navigation Satellite System (GNSS) signal; acquiring photography information for positioning of a target object photographed by a drone while the drone is flying in a GNSS shadow area from the drone equipped with a survey prism that reflects light waves from the surveying device; calculating absolute coordinates of the target object corresponding to the surveying device based on reflective metering of the survey prism and the photography information for positioning; and calculating GNSS based second location information as location information of the target object using the absolute coordinates and the first location information.

Abstract: A method for preparing isopropyl alcohol is provided. The method includes passing reaction water through a first heat exchanger and a second heat exchanger to heat the reaction water, supplying the reaction water having passed through the second heat exchanger to a reactor as a feed stream with a propylene monomer to produce a gaseous reaction product including isopropyl alcohol (IPA), purifying isopropyl alcohol from the gaseous reaction product and recovering process water, and passing the process water through the second heat exchanger to cool the process water, and transferring a portion of the cooled process water to the first heat exchanger, wherein the reaction water is brought into contact with the portion of the cooled process water in the first heat exchanger for primary heating, and then brought into contact with the process water in the second heat exchanger for secondary heating.

Abstract: An over-soldering prevented tabbing device for manufacturing solar cell modules is disclosed in which the over-soldering is prevented from occurring in the tabbing device for performing the soldering process in manufacturing the solar cell modules. The over-soldering prevented tabbing device according to an embodiment of the present disclosure includes a solar cell module cell that is transported on a conveyor by a unit of cell; a soldering head installed above the solar cell module cell and having a built-in heat source that provides heat to solder the flux coated on the transported solar cell module cell; a driver reciprocating the soldering head in a direction in which the conveyor moves; and a controller controlling the driver.

Abstract: A muscular strength assisting apparatus includes a muscular strength assisting portion that assists a user's muscular strength and a first connecting link including one side connected to one side of the muscular strength assisting portion and an opposite side rotatable relative to a shoulder of the user about a first rotation axis. The muscular strength assisting portion includes an upper arm module that assists muscular strength of an upper arm of the user and a link assembly including one side connected to one side of the upper arm module to be rotatable about a second rotation axis non-parallel to the first rotation axis and an opposite side disposed adjacent to a portion of a body of the user to be supported on the user.

Abstract: An audio signal encoding and decoding method using a neural network model, and an encoder and decoder for performing the same are disclosed. A method of encoding an audio signal using a neural network model, the method may include identifying an input signal, generating a quantized latent vector by inputting the input signal into a neural network model encoding the input signal, and generating a bitstream corresponding to the quantized latent vector, wherein the neural network model may include i) a feature extraction layer generating a latent vector by extracting a feature of the input signal, ii) a plurality of downsampling blocks downsampling the latent vector, and iii) a plurality of quantization blocks performing quantization of a downsampled latent vector.

Abstract: The present disclosure relates to a novel human antibody or an antigen-binding fragment thereof that recognizes the SMO (smoothened) protein as an antigen and binds specifically thereto. The antibody of the present disclosure binds specifically to the SMO protein, which is directly involved in major cancer progression processes such as proliferation, migration, invasion and metastasis of cancer cells, and blocks the progression those processes. Therefore, it can be usefully used as a therapeutic agent for various cancers such as colon cancer, etc., resistant cancer, metastatic cancer, or recurrent cancer.