About this Research Topic

Manuscript Submission Deadline 22 August 2022
Manuscript Extension Submission Deadline 22 September 2022

Music Source Separation (MSS) and Automatic Music Transcription (AMT) are fundamental problems in Music Information Retrieval (MIR). MSS is the task to separate individual sources from a music audio mixture. AMT is the task to transcribe music recordings into symbolic representations.

In recent ...

Music Source Separation (MSS) and Automatic Music Transcription (AMT) are fundamental problems in Music Information Retrieval (MIR). MSS is the task to separate individual sources from a music audio mixture. AMT is the task to transcribe music recordings into symbolic representations.

In recent years, MSS and AMT have attracted greater interest from both industry and academia, as they are shown to be helpful for many downstream music information retrieval tasks, including but not limited to music tagging, music generation, pitch correction, music remixing, and music performance analysis. While various approaches were explored in MSS and AMT research in the past decades, Deep-Learning based methods have shown their great promises in addressing MSS and AMT problems in large-scale, complex, and real-world scenarios.

With the rapid development of Deep-Learning research, such as representation learning, unsupervised and self-supervised learning, transfer learning, few-shot learning, meta-learning, multimodal learning, and physical-model-guided learning, many new and open questions arise for its application in MSS and AMT.

This Research Topic welcomes new perspectives, problems, and approaches for Deep-Learning based MSS and AMT research. The goal is to gather novel research ideas for Deep-Learning based MSS and AMT research and to raise its awareness in the Signal Processing community in general.

Topics of interest of this Research Topic include but are not limited to:

• New perspectives on MSS and AMT such as new definitions of sources, separation, and transcription
• New problem setups for MSS and AMT such as audio-visual separation and transcription
• Novel deep architectures, models, and approaches
• Unsupervised, self-supervised, semi-supervised learning for MSS and AMT
• Transfer learning, multi-task learning, and learning with weak labels for MSS and AMT
• Multi-instrument, multi-timbre MSS and AMT systems
• MSS and AMT for non-Western music
• Computationally efficient systems for MSS and AMT
• New datasets and evaluation metrics and methods

Topic Editor Qiuqiang Kong is currently employed by the company ByteDance Ltd. Topic Editor Zhiyao Duan is a co-founder of Mango Future Group Ltd. and the president of Mango Future America Inc. He was also employed by Kwai Inc. from June 2020 to January 2022 during his sabbatical leave from the University of Rochester. The remaining Topic Editor declares that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Keywords: Music Source Separation, Automatic Music Transcription, Signal Processing, Deep Learning


Important Note: All contributions to this Research Topic must be within the scope of the section and journal to which they are submitted, as defined in their mission statements. Frontiers reserves the right to guide an out-of-scope manuscript to a more suitable section or journal at any stage of peer review.

Topic Editors

Loading..

Topic Coordinators

Loading..

Recent Articles

Loading..

articles

Sort by:

Loading..

authors

Loading..

views

total views article views article downloads topic views

}
 
Top countries
Top referring sites
Loading..

Share on

About Frontiers Research Topics

With their unique mixes of varied contributions from Original Research to Review Articles, Research Topics unify the most influential researchers, the latest key findings and historical advances in a hot research area! Find out more on how to host your own Frontiers Research Topic or contribute to one as an author.