Dae-Young Jang

TL;DR: The main goal of the work is to propose a SER method based on concatenated CNNs and RNNs without using any traditional hand-crafted features, which was verified to have better accuracy than that achieved using conventional classification methods.

TL;DR: In this article, an apparatus and method for coding and decoding a multi-object audio signal is presented, which includes a downmixer for down-mixing the audio signals into one down-mixed audio signal and extracting supplementary information including header information and spatial cue information.

TL;DR: In this article, an object-based 3D audio system is presented, where an audio encoding unit encodes 3D information and object signals of the audio scene to transmit them through a medium and an audio decoding unit receives the encoded data through the medium, and decodes the same.

TL;DR: In this paper, the authors proposed an apparatus and method for encoding and decoding using alternative transform units according to the correlation of residual signals, which includes a first transforming unit for performing discrete cosine transform (DCT), first quantization, first inverse quantization and inverse DCT on a block basis onto residual coefficients generated after intra frame prediction or inter frame prediction.

TL;DR: In this paper, the authors proposed a method and apparatus for controlling rendering of multi-object or multi-channel audio signals based on spatial cues in a process of decoding the multichannel audio signals.