Jaehwa Chung

Minseok Kim, Woosung Choi, Jaehwa Chung et al.

Recently, many methods based on deep learning have been proposed for music source separation. Some state-of-the-art methods have shown that stacking many layers with many skip connections improve the SDR performance. Although such a deep and complex architecture shows outstanding performance, it usually requires numerous computing resources and time for training and evaluation. This paper proposes a two-stream neural network for music demixing, called KUIELab-MDX-Net, which shows a good balance of performance and required resources. The proposed model has a time-frequency branch and a time-domain branch, where each branch separates stems, respectively. It blends results from two streams to generate the final estimation. KUIELab-MDX-Net took second place on leaderboard A and third place on leaderboard B in the Music Demixing Challenge at ISMIR 2021. This paper also summarizes experimental results on another benchmark, MUSDB18. Our source code is available online.

Jinsung Kim, Yeong-Seok Jeong, Woosung Choi et al.

In recent years, neural network based methods have been proposed as a method that cangenerate representations from music, but they are not human readable and hardly analyzable oreditable by a human. To address this issue, we propose a novel method to learn source-awarelatent representations of music through Vector-Quantized Variational Auto-Encoder(VQ-VAE).We train our VQ-VAE to encode an input mixture into a tensor of integers in a discrete latentspace, and design them to have a decomposed structure which allows humans to manipulatethe latent vector in a source-aware manner. This paper also shows that we can generate basslines by estimating latent vectors in a discrete space.

Yeong-Seok Jeong, Jinsung Kim, Woosung Choi et al.

Conditioned source separations have attracted significant attention because of their flexibility, applicability and extensionality. Their performance was usually inferior to the existing approaches, such as the single source separation model. However, a recently proposed method called LaSAFT-Net has shown that conditioned models can show comparable performance against existing single-source separation models. This paper presents LightSAFT-Net, a lightweight version of LaSAFT-Net. As a baseline, it provided a sufficient SDR performance for comparison during the Music Demixing Challenge at ISMIR 2021. This paper also enhances the existing LightSAFT-Net by replacing the LightSAFT blocks in the encoder with TFC-TDF blocks. Our enhanced LightSAFT-Net outperforms the previous one with fewer parameters.Conditioned source separations have attracted significant attention because of their flexibility, applicability and extensionality. Their performance was usually inferior to the existing approaches, such as the single source separation model. However, a recently proposed method called LaSAFT-Net has shown that conditioned models can show comparable performance against existing single-source separation models. This paper presents LightSAFT-Net, a lightweight version of LaSAFT-Net. As a baseline, it provided a sufficient SDR performance for comparison during the Music Demixing Challenge at ISMIR 2021.

Woosung Choi, Minseok Kim, Marco A. Martínez Ramírez et al.

This paper proposes a neural network that performs audio transformations to user-specified sources (e.g., vocals) of a given audio track according to a given description while preserving other sources not mentioned in the description. Audio Manipulation on a Specific Source (AMSS) is challenging because a sound object (i.e., a waveform sample or frequency bin) is `transparent'; it usually carries information from multiple sources, in contrast to a pixel in an image. To address this challenging problem, we propose AMSS-Net, which extracts latent sources and selectively manipulates them while preserving irrelevant sources. We also propose an evaluation benchmark for several AMSS tasks, and we show that AMSS-Net outperforms baselines on several AMSS tasks via objective metrics and empirical verification.

Woosung Choi, Minseok Kim, Jaehwa Chung et al.

Recent deep-learning approaches have shown that Frequency Transformation (FT) blocks can significantly improve spectrogram-based single-source separation models by capturing frequency patterns. The goal of this paper is to extend the FT block to fit the multi-source task. We propose the Latent Source Attentive Frequency Transformation (LaSAFT) block to capture source-dependent frequency patterns. We also propose the Gated Point-wise Convolutional Modulation (GPoCM), an extension of Feature-wise Linear Modulation (FiLM), to modulate internal features. By employing these two novel methods, we extend the Conditioned-U-Net (CUNet) for multi-source separation, and the experimental results indicate that our LaSAFT and GPoCM can improve the CUNet's performance, achieving state-of-the-art SDR performance on several MUSDB18 source separation tasks.

Woosung Choi, Minseok Kim, Jaehwa Chung et al.

Singing Voice Separation (SVS) tries to separate singing voice from a given mixed musical signal. Recently, many U-Net-based models have been proposed for the SVS task, but there were no existing works that evaluate and compare various types of intermediate blocks that can be used in the U-Net architecture. In this paper, we introduce a variety of intermediate spectrogram transformation blocks. We implement U-nets based on these blocks and train them on complex-valued spectrograms to consider both magnitude and phase. These networks are then compared on the SDR metric. When using a particular block composed of convolutional and fully-connected layers, it achieves state-of-the-art SDR on the MUSDB singing voice separation task by a large margin of 0.9 dB. Our code and models are available online.