Abstract: A bitstream reconstruction method and apparatus are provided. The method includes constructing an initial bitstream by rendering sound source information and geometry information within a reference radius from an initial location of a user accessing a virtual space into spatial audio, collecting location information according to a movement of the user within the virtual space, and reconstructing, based on a relationship between the reference radius and a movement radius identified according to the collected location information, the initial bitstream constructed by corresponding to the initial location of the user.

Abstract: A method and apparatus for generating late reverberation are provided. The method includes generating a reverberation parameter required to generate late reverberation based on an early room impulse response, outputting late reverberation based on the reverberation parameter, and outputting a room impulse response based on the early room impulse response and the late reverberation.

Abstract: A rendering method of an object-based audio signal and an apparatus for performing the same are provided. The rendering method of an object-based audio signal includes obtaining a rendered audio signal, performing clipping prevention on the rendered audio signal using a first limiter, mixing a signal output by the first limiter using a mixer, and performing clipping prevention on the mixed signal using a second limiter.

Abstract: Disclosed is a method and apparatus for label encoding in a multi-sound event interval. The method includes identifying an event interval in which a plurality of sound events occurs in a sound signal, separating a sound source into sound event signals corresponding to each sound event by performing sound source separation on the event interval, determining energy information for each of the sound event signals, and performing label encoding based on the energy information.

Abstract: Disclosed are a training method for a learning model for recognizing an acoustic signal, a method of recognizing an acoustic signal using the learning model, and devices for performing the methods. The method of recognizing an acoustic signal using a learning model includes identifying an acoustic signal including an acoustic event or acoustic scene, determining an acoustic feature of the acoustic signal, dividing the determined acoustic feature for each of a plurality of frequency band intervals, and determining the acoustic event or acoustic scene included in the acoustic signal by inputting the divided acoustic features to a trained learning model.

Abstract: A rendering method of an object-based audio signal and an apparatus for performing the same are provided. The rendering method of an object-based audio signal includes obtaining a rendered audio signal, performing clipping prevention on the rendered audio signal using a first limiter, mixing a signal output by the first limiter using a mixer, and performing clipping prevention on the mixed signal using a second limiter.

Abstract: A method of rendering object-based audio and an electronic device performing the method are provided. The method includes identifying a bitstream, determining a reference distance of an object sound source based on the bitstream, determining a minimum distance for applying distance-dependent attenuation, based on the reference distance, and determining a gain of object-based audio included in the bitstream based on the reference distance and the minimum distance.

Abstract: A method of rendering object-based audio and an electronic device performing the method are disclosed. The method includes identifying metadata of object-based audio, identifying an audio source distance between the object-based audio and a listener using the metadata, determining a minimum distance of the object-based audio to apply attenuation according to the audio source distance, based on a reference distance of the object-based audio in the metadata, and rendering the object-based audio using the audio source distance and the minimum distance.

Abstract: A method of rendering object-based audio and an electronic device for performing the method are disclosed. The method includes identifying metadata of the object-based audio, determining whether the metadata includes a parameter set for an atmospheric absorption effect for each distance, and rendering the object-based audio, using a distance between the object-based audio and a listener obtained using the metadata and the atmospheric absorption effect according to an effect of a medium attenuation based on the parameter, when the metadata includes the parameter.

Abstract: A method of rendering object audio and electronic device for performing the method are provided. The method includes identifying object audio and metadata including a reference distance of the object audio and an audio source distance between the object audio and a listener, identifying a maximum distance of the object audio determined based on the reference distance, when the audio source distance is less than or equal to the maximum distance, determining a gain of the object audio based on the audio source distance and the reference distance, and rendering the object audio according to the gain of the object audio.

Abstract: Provided are a method of training a neural network model, a method of recognizing an acoustic event and an acoustic direction, and an electronic device for performing the methods. A method of training a neural network model according to an example embodiment includes generating a heatmap indicating an acoustic event and an acoustic direction in which the acoustic event occurs by using training data, outputting a result of recognizing the acoustic event and the acoustic direction by inputting a feature extracted using the training data into a neural network model for recognizing the acoustic event and the acoustic direction of the training data, and training the neural network model by using the result and the heatmap.

Abstract: Provided are a method of recognizing sound, a method of training a sound recognition model, and an electronic device performing the same methods. A method of training a sound recognition model according to an example embodiment may include converting training data labeled with a sound class into a feature vector, storing the feature vector in a feature queue, transferring the feature vector stored in the feature queue to a block queue according to an operation of a feature vector transfer timer, inputting the feature vector of the block queue into a sound recognition model trained to predict the sound class and storing an output result in a result queue, transferring the feature vector stored in the feature queue corresponding to timing at which the result is output to the block queue by the feature vector transfer timer when the result is output.

Abstract: Disclosed are methods of training a deep learning model and predicting a class and an electronic device for performing the methods. A method of training a deep learning model may include identifying training data labeled for each class, determining whether to augment the training data based on overall recognition performance indicating prediction accuracy of the deep learning model calculated in a previous epoch, augmenting the training data based on class-specific recognition performance indicating class-specific prediction accuracy of the deep learning model calculated in the previous epoch, predicting a class by inputting the training data or the training data that is augmented to the deep learning model according to a determination of whether to augment the training data, and training the deep learning model based on a labeled class and the predicted class.

Abstract: Disclosed is an apparatus and method for augmenting training data using a notch filter. The method may include obtaining original data, and obtaining training data having a modified frequency component from the original data by filtering the original data using a filter configured to remove a component of a predetermined frequency band.

Abstract: Disclosed are methods of training an acoustic scene classification model and classifying an acoustic scene and an electronic device for performing the methods. The training method of an acoustic scene classification model includes inputting training data labeled as an acoustic scene to the acoustic scene classification model that is repeatedly trained by using the training data and outputting a first result predicting the acoustic scene, updating the weight of the auxiliary model configured to induce training of the acoustic scene classification model, based on a weight of the acoustic scene classification model and a weight of an auxiliary model in a previous epoch, inputting the training data to the auxiliary model and outputting a second result, calculating a cost function, based on the first result, the second result, and labeling of acoustic data, and updating the weight of the acoustic scene classification model, based on the cost function.

Abstract: Disclosed is a method and apparatus for label encoding in a multi-sound event interval. The method includes identifying an event interval in which a plurality of sound events occurs in a sound signal, separating a sound source into sound event signals corresponding to each sound event by performing sound source separation on the event interval, determining energy information for each of the sound event signals, and performing label encoding based on the energy information.

Abstract: Disclosed are a training method for a learning model for recognizing an acoustic signal, a method of recognizing an acoustic signal using the learning model, and devices for performing the methods. The method of recognizing an acoustic signal using a learning model includes identifying an acoustic signal including an acoustic event or acoustic scene, determining an acoustic feature of the acoustic signal, dividing the determined acoustic feature for each of a plurality of frequency band intervals, and determining the acoustic event or acoustic scene included in the acoustic signal by inputting the divided acoustic features to a trained learning model.

Abstract: Provided is a sound event recognition method that may improve a sound event recognition performance using a correlation between difference sound signal feature parameters based on a neural network, in detail, that may extract a sound signal feature parameter from a sound signal including a sound event, and recognize the sound event included in the sound signal by applying a convolutional neural network (CNN) trained using the sound signal feature parameter.

Abstract: A substrate polishing system includes: a polishing machine and a substrate transporter. The polishing machine includes: a lower surface plate to which a substrate is mounted, and an upper surface plate which faces the lower surface plate and polishes the substrate in cooperation with the lower surface plate, the upper surface plate having a larger area than the substrate mounted on the lower surface plate. The substrate transporter is adjacent to the polishing machine and commonly transports the substrate to and from the polishing machine in a first direction, attaches the substrate to the lower surface plate before polishing thereof, and separates from the lower surface plate the substrate after polishing thereof.

Abstract: Disclosed is an apparatus and method for augmenting training data using a notch filter. The method may include obtaining original data, and obtaining training data having a modified frequency component from the original data by filtering the original data using a filter configured to remove a component of a predetermined frequency band.