Music Foundation Model as Generic Booster for Music Downstream Tasks

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Liao, WeiHsiang, Takida, Yuhta, Ikemiya, Yukara, Zhong, Zhi, Lai, Chieh-Hsin, Fabbro, Giorgio, Shimada, Kazuki, Toyama, Keisuke, Cheuk, Kinwai, Martínez-Ramírez, Marco A., Takahashi, Shusuke, Uhlich, Stefan, Akama, Taketo, Choi, Woosung, Koyama, Yuichiro, Mitsufuji, Yuki
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913860260200448
author Liao, WeiHsiang
Takida, Yuhta
Ikemiya, Yukara
Zhong, Zhi
Lai, Chieh-Hsin
Fabbro, Giorgio
Shimada, Kazuki
Toyama, Keisuke
Cheuk, Kinwai
Martínez-Ramírez, Marco A.
Takahashi, Shusuke
Uhlich, Stefan
Akama, Taketo
Choi, Woosung
Koyama, Yuichiro
Mitsufuji, Yuki
author_facet Liao, WeiHsiang
Takida, Yuhta
Ikemiya, Yukara
Zhong, Zhi
Lai, Chieh-Hsin
Fabbro, Giorgio
Shimada, Kazuki
Toyama, Keisuke
Cheuk, Kinwai
Martínez-Ramírez, Marco A.
Takahashi, Shusuke
Uhlich, Stefan
Akama, Taketo
Choi, Woosung
Koyama, Yuichiro
Mitsufuji, Yuki
contents We demonstrate the efficacy of using intermediate representations from a single foundation model to enhance various music downstream tasks. We introduce SoniDo, a music foundation model (MFM) designed to extract hierarchical features from target music samples. By leveraging hierarchical intermediate features, SoniDo constrains the information granularity, leading to improved performance across various downstream tasks including both understanding and generative tasks. We specifically evaluated this approach on representative tasks such as music tagging, music transcription, music source separation, and music mixing. Our results reveal that the features extracted from foundation models provide valuable enhancements in training downstream task models. This highlights the capability of using features extracted from music foundation models as a booster for downstream tasks. Our approach not only benefits existing task-specific models but also supports music downstream tasks constrained by data scarcity. This paves the way for more effective and accessible music processing solutions.
format Preprint
id arxiv_https___arxiv_org_abs_2411_01135
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Music Foundation Model as Generic Booster for Music Downstream Tasks
Liao, WeiHsiang
Takida, Yuhta
Ikemiya, Yukara
Zhong, Zhi
Lai, Chieh-Hsin
Fabbro, Giorgio
Shimada, Kazuki
Toyama, Keisuke
Cheuk, Kinwai
Martínez-Ramírez, Marco A.
Takahashi, Shusuke
Uhlich, Stefan
Akama, Taketo
Choi, Woosung
Koyama, Yuichiro
Mitsufuji, Yuki
Sound
Information Retrieval
Machine Learning
Audio and Speech Processing
We demonstrate the efficacy of using intermediate representations from a single foundation model to enhance various music downstream tasks. We introduce SoniDo, a music foundation model (MFM) designed to extract hierarchical features from target music samples. By leveraging hierarchical intermediate features, SoniDo constrains the information granularity, leading to improved performance across various downstream tasks including both understanding and generative tasks. We specifically evaluated this approach on representative tasks such as music tagging, music transcription, music source separation, and music mixing. Our results reveal that the features extracted from foundation models provide valuable enhancements in training downstream task models. This highlights the capability of using features extracted from music foundation models as a booster for downstream tasks. Our approach not only benefits existing task-specific models but also supports music downstream tasks constrained by data scarcity. This paves the way for more effective and accessible music processing solutions.
title Music Foundation Model as Generic Booster for Music Downstream Tasks
topic Sound
Information Retrieval
Machine Learning
Audio and Speech Processing
url https://arxiv.org/abs/2411.01135