Abstract: A device for AI-based vehicle diagnosis using CAN data may include an engine; a vibration sensor mounted in an engine compartment in which the engine is mounted and configured for detecting a vibration signal; and a controller area network (CAN) communicating with one or more of an environmental condition, a vehicle status, an engine status, and an engine control parameter, wherein data preprocessing from the vibration sensor and the CAN is performed to determine features in which correlation between vibration data (dB) exceeding a threshold value of irregular vibrations being generated by the engine and the CAN data is equal to or greater than 90%.

Abstract: Disclosed are a method and an apparatus for detecting a voice end point by using acoustic and language modeling information to accomplish strong voice recognition. A voice end point detection method according to an embodiment may comprise the steps of: inputting an acoustic feature vector sequence extracted from a microphone input signal into an acoustic embedding extraction unit, a phonemic embedding extraction unit, and a decoder embedding extraction unit, which are based on a recurrent neural network (RNN); combining acoustic embedding, phonemic embedding, and decoder embedding to configure a feature vector by the acoustic embedding extraction unit, the phonemic embedding extraction unit, and the decoder embedding extraction unit; and inputting the combined feature vector into a deep neural network (DNN)-based classifier to detect a voice end point.

Abstract: Presented are a combined learning method and device using a transformed loss function and feature enhancement based on a deep neural network for speaker recognition that is robust in a noisy environment. A combined learning method using a transformed loss function and feature enhancement based on a deep neural network, according to one embodiment, can comprise the steps of: learning a feature enhancement model based on a deep neural network; learning a speaker feature vector extraction model based on the deep neural network; connecting an output layer of the feature enhancement model with an input layer of the speaker feature vector extraction model; and considering the connected feature enhancement model and speaker feature vector extraction model as one mode and performing combined learning for additional learning.

Abstract: Presented are a combined learning method and device using a transformed loss function and feature enhancement based on a deep neural network for speaker recognition that is robust to a noisy environment.

Abstract: A method for rearranging image cuts of cartoon content is performed by a computing device and includes the steps of loading first content in which a plurality of image cuts are arrayed two-dimensionally; extracting a plurality of cut areas, in which the plurality of image cuts from the first content are positioned, respectively; determining the arrayed order of the plurality of image cuts; and generating second content by rearranging the plurality of cut areas according to the arrayed order.

Abstract: Proposed are a deep neural network-based non-autoregressive voice synthesizing method and a system therefor. A deep neural network-based non-autoregressive voice synthesizing system according to an embodiment may comprise: a voice feature vector column synthesizing unit which constitutes a non-recursive deep neural network based on multiple decoders, and gradually produces a voice feature vector column through the multiple decoders from a template including temporal information of a voice; and a voice reconstituting unit which transforms the voice feature vector column into voice data, wherein the voice feature vector column synthesizing unit produces a template input, and produces a voice feature vector column by adding, to the template input, sentence data refined through an attention mechanism.

Abstract: A deep learning-based coloring system includes a memory network configured to provide a color feature in response to a specific query and a coloring network configured to perform coloring, based on the color feature generated by the memory network. The memory network includes: a query generation unit configured to generate a query; a neighbor calculation unit configured to calculate k-nearest neighbors, based on similarities between the query and key memory values; a color feature determination unit configured to generate color features for indicating color information stored in the key memory; a threshold triplet loss calculation unit configured to calculate a threshold triplet loss, based on a comparison between a threshold and a distance between the color features; and a memory update unit configured to update a memory, based on whether a distance between a top value and a value of a newly input query is within the threshold.

Abstract: Proposed are a generative adversarial network-based speech bandwidth extender and extension method. A generative adversarial network-based speech bandwidth extension method, according to an embodiment, comprises the steps of: extracting feature vectors from a narrowband (NB) signal and a wideband (WB) signal of a speech; estimating the feature vector of the wideband signal from the feature vector of the narrowband signal; and learning a deep neural network classification model for discriminating the estimated feature vector of the wideband signal from the actually extracted feature vector of the wideband signal and the actually extracted feature vector of the narrowband signal.

Abstract: An apparatus for and a method for estimating blood pressure are provided. The apparatus for estimating blood pressure includes: a sensor configured to measure a pulse wave signal from an object; and a processor configured to obtain a mean arterial pressure (MAP) based on the pulse wave signal, configured to classify a phase of the obtained MAP according to at least one classification criterion, and to obtain a systolic blood pressure (SBP) by using an estimation model corresponding to the classified phase of the MAP among estimation models corresponding to respective phases of the MAP.

Abstract: Disclosed herein are an apparatus and method for recognizing a voice speaker. The apparatus for recognizing a voice speaker includes a voice feature extraction unit configured to extract a feature vector from a voice signal inputted through a microphone; and a speaker recognition unit configured to calculate a speaker recognition score by selecting a reverberant environment from multiple reverberant environment learning data sets based on the feature vector extracted by the voice feature extraction unit and to recognize a speaker by assigning a weight depending on the selected reverberant environment to the speaker recognition score.

Abstract: A camera system with a complementary pixlet structure and a method of operating the same are provided. The camera system includes an image sensor that includes at least one 2×2 pixel block including a first pixel, a second pixel, and two third pixels—the two third pixels are disposed at positions diagonal to each other in the 2×2 pixel block and include deflected small pixlets, which are deflected in opposite directions to be symmetrical to each other with respect to each pixel center, and large pixlets adjacent to the deflected small pixlets, respectively, and each pixlet includes an photodiode converting an optical signal to an electrical signal and a depth calculator that receives images acquired from the deflected small pixlets of the two third pixels and calculates a depth between the image sensor and an object using a parallax between the images.

Abstract: A camera system with a complementary pixlet structure and a method of operating the same are provided. The camera system includes an image sensor that includes at least one 2×2 pixel block including a first pixel, a second pixel, and two third pixels—the two third pixels are disposed at positions diagonal to each other in the 2×2 pixel block and include deflected small pixlets, which are deflected in opposite directions to be symmetrical to each other with respect to each pixel center, and large pixlets adjacent to the deflected small pixlets, respectively, and each pixlet includes an photodiode converting an optical signal to an electrical signal and a depth calculator that receives images acquired from the deflected small pixlets of the two third pixels and calculates a depth between the image sensor and an object using a parallax between the images.

Abstract: A camera system with a complementary pixlet structure and a method of operating the same are provided. According to an embodiment, the camera system includes an image sensor that includes two pixels, each of the two pixels including a deflected small pixlet deflected in one direction based on a pixel center and a large pixlet disposed adjacent to the deflected small pixlet, each pixlet including a photodiode converting an optical signal to an electric signal, and the deflected small pixlets of the two pixels being arranged to be symmetrical to each other with respect to each of the pixel centers within each of the two pixels, respectively, and a depth calculator that receives images acquired from the deflected small pixlets of the two pixels and calculates a depth between the image sensor and an object using a parallax between the images.

Abstract: Proposed are a generative adversarial network-based speech bandwidth extender and extension method. A generative adversarial network-based speech bandwidth extension method, according to an embodiment, comprises the steps of: extracting feature vectors from a narrowband (NB) signal and a wideband (WB) signal of a speech; estimating the feature vector of the wideband signal from the feature vector of the narrowband signal; and learning a deep neural network classification model for discriminating the estimated feature vector of the wideband signal from the actually extracted feature vector of the wideband signal and the actually extracted feature vector of the narrowband signal.

Abstract: Disclosed is a deep neural network-based method and apparatus for combining noise and echo removal. The deep neural network-based method for combining noise and echo removal according to one embodiment of the present invention may comprise the steps of extracting a feature vector from an audio signal that includes noise and echo; and acquiring a final audio signal from which both noise and echo have been removed, by using a combined nose and echo removal gain estimated by means of the feature vector and deep neural network DNN.

Abstract: A camera system includes a single lens and an image sensor including a reference pixel array including a plurality of W (white) pixels in a two-dimensional arrangement and a single microlens formed on the plurality of W pixels to be shared, and at least one color pixel array including two W pixels and two color pixels in a two-dimensional arrangement, and a single microlens disposed on the two W pixels and the two color pixels to be shared. Light shielding layers formed with Offset Pixel Apertures (OPAs) are disposed on the plurality of W pixels included in the reference pixel array and the two W pixels included in the at least one color pixel array, respectively, and the OPAs are formed on the light shielding layers in the reference pixel array and the at least one color pixel array, respectively, to maximize a spaced distance between the OPAs.

Abstract: A camera system is provided to increase a baseline. The camera system includes a single lens, and an image sensor includes at least one pixel array, each of the at least one pixel array including a plurality of pixels in a two-dimensional arrangement and a single microlens disposed on the plurality of pixels to be shared. Light shielding layers formed with Offset Pixel Apertures (OPAs) are disposed on at least two pixels of the plurality of pixels, and the OPAs are formed on the light shielding layers to maximize a spaced distance between the OPAs.

Abstract: Disclosed are a packet loss concealment method and apparatus a using a generative adversarial network. A method for packet loss concealment in voice communication may include training a classification model based on a generative adversarial network (GAN) with respect to a voice signal including a plurality of frames, training a generative model having a contention relation with the classification model based on the GAN, estimating lost packet information based on the trained generative model with respect to the voice signal encoded by a codec, and restoring a lost packet based on the estimated packet information.

Abstract: A multichannel microphone-based reverberation time estimation method and device which use a deep neural network (DNN) are disclosed. A multichannel microphone-based reverberation time estimation method using a DNN, according to one embodiment, comprises the steps of: receiving a voice signal through a multichannel microphone; deriving a feature vector including spatial information by using the inputted voice signal; and estimating the degree of reverberation by applying the feature vector to the DNN.

Abstract: A noise data artificial intelligence learning method for identifying the source of problematic noise may include a noise data pre-conditioning method for identifying the source of problematic noise including: selecting a unit frame for the problematic noise among noises sampled with time; dividing the unit frame into N segments; analyzing frequency characteristic for each segment of the N segments and extracting a frequency component of each segment by applying Log Mel Filter; and outputting a feature parameter as one representative frame by averaging information on the N segments, wherein an artificial intelligence learning by the feature parameter extracted according to a change in time by the noise data pre-conditioning method applies Bidirectional RNN.