MedleyVox: An Evaluation Dataset for Multiple Singing Voices Separation

TL;DR

MedleyVox introduces a new dataset and baseline methods for separating multiple singing voices, addressing a key gap in music source separation research and enabling future advancements in the field.

Contribution

The paper presents MedleyVox, a novel dataset for multiple singing voices separation, along with an improved super-resolution network and a strategy for constructing multi-singing mixtures.

Findings

iSRNet enhances separation quality significantly.

Joint training with Conv-TasNet yields competitive results.

MedleyVox enables standardized evaluation of multi-singing separation.

Abstract

Separation of multiple singing voices into each voice is a rarely studied area in music source separation research. The absence of a benchmark dataset has hindered its progress. In this paper, we present an evaluation dataset and provide baseline studies for multiple singing voices separation. First, we introduce MedleyVox, an evaluation dataset for multiple singing voices separation. We specify the problem definition in this dataset by categorizing it into i) unison, ii) duet, iii) main vs. rest, and iv) N-singing separation. Second, to overcome the absence of existing multi-singing datasets for a training purpose, we present a strategy for construction of multiple singing mixtures using various single-singing datasets. Third, we propose the improved super-resolution network (iSRNet), which greatly enhances initial estimates of separation networks. Jointly trained with the Conv-TasNet and the multi-singing mixture construction strategy, the proposed iSRNet achieved comparable performance to ideal time-frequency masks on duet and unison subsets of MedleyVox. Audio samples, the dataset, and codes are available on our website (https://github.com/jeonchangbin49/MedleyVox).