---
license: cc-by-nc-4.0
language:
- ja
tags:
- music
- speech
---

# MRDLA trained with the jaCappella corpus for vocal ensemble separation

This model was trained by Tomohiko Nakamura using [the codebase](https://github.com/TomohikoNakamura/asteroid_jaCappella)).
It was trained on the vocal ensemble separation task of [the jaCappella dataset](https://tomohikonakamura.github.io/jaCappella_corpus/).

# License
See [the jaCappella dataset page](https://tomohikonakamura.github.io/jaCappella_corpus/).

# Citation
See [the jaCappella dataset page](https://tomohikonakamura.github.io/jaCappella_corpus/).

For MRDLA, please cite the following paper.
```
@article{TNakamura202104IEEEACMTASLP,
 author={Nakamura, Tomohiko and Kozuka, Shihori and Saruwatari, Hiroshi},
 journal = {IEEE/ACM Transactions on Audio, Speech, and Language Processing},
 title = {Time-domain audio source separation with neural networks based on multiresolution analysis},
 year=2021,
 doi={10.1109/TASLP.2021.3072496},
 month=apr,
 volume=29,
 pages={1687--1701},
}
```