Vocal separation using improved robust principal component analysis and post-processing

Tomohiro Watanabe, Takanori Fujisawa, Masaaki Ikehara

Research output: Chapter in Book/Report/Conference proceedingConference contribution

1 Citation (Scopus)

Abstract

Due to the spread of the Internet and improvement of the audio compression technology, we need technologies to handle a large number of audio files. Separating singing voice from music is one of them and it is used in many applications, such as lyric retrieval and singer recognition. Recently, robust principal component analysis (RPCA) has been proposed, which makes use of repetition of a phrase in an accompaniment. We assume that there are some points to be improved in RPCA. In this paper, we propose a developed method based on the RPCA algorithm. First, we propose the extended RPCA algorithm that makes use of the two features of the spectrogram. Second, we apply simple post-processing to remove noise effectively. Lastly, we experiment our algorithm using the MIR-1K dataset and confirm that the proposed algorithm shows better separation performance than the conventional RPCA method.

Original languageEnglish
Title of host publication2016 IEEE 59th International Midwest Symposium on Circuits and Systems, MWSCAS 2016
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9781509009169
DOIs
Publication statusPublished - 2017 Mar 2
Event59th IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2016 - Abu Dhabi, United Arab Emirates
Duration: 2016 Oct 162016 Oct 19

Other

Other59th IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2016
CountryUnited Arab Emirates
CityAbu Dhabi
Period16/10/1616/10/19

Fingerprint

Principal component analysis
Processing
Internet
Experiments

ASJC Scopus subject areas

  • Electronic, Optical and Magnetic Materials
  • Electrical and Electronic Engineering

Cite this

Watanabe, T., Fujisawa, T., & Ikehara, M. (2017). Vocal separation using improved robust principal component analysis and post-processing. In 2016 IEEE 59th International Midwest Symposium on Circuits and Systems, MWSCAS 2016 [7870055] Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/MWSCAS.2016.7870055

Vocal separation using improved robust principal component analysis and post-processing. / Watanabe, Tomohiro; Fujisawa, Takanori; Ikehara, Masaaki.

2016 IEEE 59th International Midwest Symposium on Circuits and Systems, MWSCAS 2016. Institute of Electrical and Electronics Engineers Inc., 2017. 7870055.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Watanabe, T, Fujisawa, T & Ikehara, M 2017, Vocal separation using improved robust principal component analysis and post-processing. in 2016 IEEE 59th International Midwest Symposium on Circuits and Systems, MWSCAS 2016., 7870055, Institute of Electrical and Electronics Engineers Inc., 59th IEEE International Midwest Symposium on Circuits and Systems, MWSCAS 2016, Abu Dhabi, United Arab Emirates, 16/10/16. https://doi.org/10.1109/MWSCAS.2016.7870055
Watanabe T, Fujisawa T, Ikehara M. Vocal separation using improved robust principal component analysis and post-processing. In 2016 IEEE 59th International Midwest Symposium on Circuits and Systems, MWSCAS 2016. Institute of Electrical and Electronics Engineers Inc. 2017. 7870055 https://doi.org/10.1109/MWSCAS.2016.7870055
Watanabe, Tomohiro ; Fujisawa, Takanori ; Ikehara, Masaaki. / Vocal separation using improved robust principal component analysis and post-processing. 2016 IEEE 59th International Midwest Symposium on Circuits and Systems, MWSCAS 2016. Institute of Electrical and Electronics Engineers Inc., 2017.
@inproceedings{be8311da60ce449ab55147a753c79cfb,
title = "Vocal separation using improved robust principal component analysis and post-processing",
abstract = "Due to the spread of the Internet and improvement of the audio compression technology, we need technologies to handle a large number of audio files. Separating singing voice from music is one of them and it is used in many applications, such as lyric retrieval and singer recognition. Recently, robust principal component analysis (RPCA) has been proposed, which makes use of repetition of a phrase in an accompaniment. We assume that there are some points to be improved in RPCA. In this paper, we propose a developed method based on the RPCA algorithm. First, we propose the extended RPCA algorithm that makes use of the two features of the spectrogram. Second, we apply simple post-processing to remove noise effectively. Lastly, we experiment our algorithm using the MIR-1K dataset and confirm that the proposed algorithm shows better separation performance than the conventional RPCA method.",
author = "Tomohiro Watanabe and Takanori Fujisawa and Masaaki Ikehara",
year = "2017",
month = "3",
day = "2",
doi = "10.1109/MWSCAS.2016.7870055",
language = "English",
booktitle = "2016 IEEE 59th International Midwest Symposium on Circuits and Systems, MWSCAS 2016",
publisher = "Institute of Electrical and Electronics Engineers Inc.",

}

TY - GEN

T1 - Vocal separation using improved robust principal component analysis and post-processing

AU - Watanabe, Tomohiro

AU - Fujisawa, Takanori

AU - Ikehara, Masaaki

PY - 2017/3/2

Y1 - 2017/3/2

N2 - Due to the spread of the Internet and improvement of the audio compression technology, we need technologies to handle a large number of audio files. Separating singing voice from music is one of them and it is used in many applications, such as lyric retrieval and singer recognition. Recently, robust principal component analysis (RPCA) has been proposed, which makes use of repetition of a phrase in an accompaniment. We assume that there are some points to be improved in RPCA. In this paper, we propose a developed method based on the RPCA algorithm. First, we propose the extended RPCA algorithm that makes use of the two features of the spectrogram. Second, we apply simple post-processing to remove noise effectively. Lastly, we experiment our algorithm using the MIR-1K dataset and confirm that the proposed algorithm shows better separation performance than the conventional RPCA method.

AB - Due to the spread of the Internet and improvement of the audio compression technology, we need technologies to handle a large number of audio files. Separating singing voice from music is one of them and it is used in many applications, such as lyric retrieval and singer recognition. Recently, robust principal component analysis (RPCA) has been proposed, which makes use of repetition of a phrase in an accompaniment. We assume that there are some points to be improved in RPCA. In this paper, we propose a developed method based on the RPCA algorithm. First, we propose the extended RPCA algorithm that makes use of the two features of the spectrogram. Second, we apply simple post-processing to remove noise effectively. Lastly, we experiment our algorithm using the MIR-1K dataset and confirm that the proposed algorithm shows better separation performance than the conventional RPCA method.

UR - http://www.scopus.com/inward/record.url?scp=85015950183&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85015950183&partnerID=8YFLogxK

U2 - 10.1109/MWSCAS.2016.7870055

DO - 10.1109/MWSCAS.2016.7870055

M3 - Conference contribution

AN - SCOPUS:85015950183

BT - 2016 IEEE 59th International Midwest Symposium on Circuits and Systems, MWSCAS 2016

PB - Institute of Electrical and Electronics Engineers Inc.

ER -