# Singing voice separation: a study on training data

arXiv: https://arxiv.org/abs/1906.02618 · HTML: https://mlsanity.com/paper/1906.02618/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Jun 2019.

## Built on by (most cited first, 5 of 5)

- [Moisesdb: A dataset for source separation beyond 4-stems](https://mlsanity.com/paper/2307.15913/) (2023)
- [The Chamber Ensemble Generator: Limitless High-Quality MIR Data via Generative Modeling](https://mlsanity.com/paper/2209.14458/) (2022)
- [Benchmarks and leaderboards for sound demixing tasks](https://mlsanity.com/paper/2305.07489/) (2023)
- [Modeling the Compatibility of Stem Tracks to Generate Music Mashups](https://mlsanity.com/paper/2103.14208/) (2021)
- [Improved singing voice separation with chromagram-based pitch-aware remixing](https://mlsanity.com/paper/2203.15092/) (2022)

## Abstract (arXiv)

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.