# Singing voice separation: a study on training data

## TL;DR

This paper investigates how the characteristics of training datasets affect the performance of singing voice separation systems, emphasizing the importance of quality, diversity, and data augmentation techniques.

## Contribution

It provides a detailed analysis of dataset features impacting separation quality and offers insights into effective data augmentation methods for singing voice separation.

## Key findings

- Separation quality improves with diverse and high-quality datasets.
- Data augmentation can enhance model robustness and performance.
- Dataset characteristics are crucial for optimizing singing voice separation algorithms.

## Abstract

In the recent years, singing voice separation systems showed increased performance due to the use of supervised training. The design of training datasets is known as a crucial factor in the performance of such systems. We investigate on how the characteristics of the training dataset impacts the separation performances of state-of-the-art singing voice separation algorithms. We show that the separation quality and diversity are two important and complementary assets of a good training dataset. We also provide insights on possible transforms to perform data augmentation for this task.