Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Music2Fail: Transfer Music to Failed Recorder Style

Chon In Leong, I-Ling Chung, Kin-Fong Chao, Jun-You Wang, Yi-Hsuan Yang, Jyh-Shing Roger Jang

TL;DR

This work attempts to transfer normally played music into off-pitch recorder music, which is called ``failed-style recorder'', and study the results of the conversion.

Abstract

The goal of music style transfer is to convert a music performance by one instrument into another while keeping the musical contents unchanged. In this paper, we investigate another style transfer scenario called ``failed-music style transfer''. Unlike the usual music style transfer where the content remains the same and only the instrumental characteristics are changed, this scenario seeks to transfer the music from the source instrument to the target instrument which is deliberately performed off-pitch. Our work attempts to transfer normally played music into off-pitch recorder music, which we call ``failed-style recorder'', and study the results of the conversion. To carry out this work, we have also proposed a dataset of failed-style recorders for this task, called ``FR109 Dataset''. Such an experiment explores the music style transfer task in a more expressive setting, as the generated audio should sound like an ``off-pitch recorder'' while maintaining a certain degree of naturalness.

Music2Fail: Transfer Music to Failed Recorder Style

TL;DR

This work attempts to transfer normally played music into off-pitch recorder music, which is called ``failed-style recorder'', and study the results of the conversion.

Abstract

The goal of music style transfer is to convert a music performance by one instrument into another while keeping the musical contents unchanged. In this paper, we investigate another style transfer scenario called ``failed-music style transfer''. Unlike the usual music style transfer where the content remains the same and only the instrumental characteristics are changed, this scenario seeks to transfer the music from the source instrument to the target instrument which is deliberately performed off-pitch. Our work attempts to transfer normally played music into off-pitch recorder music, which we call ``failed-style recorder'', and study the results of the conversion. To carry out this work, we have also proposed a dataset of failed-style recorders for this task, called ``FR109 Dataset''. Such an experiment explores the music style transfer task in a more expressive setting, as the generated audio should sound like an ``off-pitch recorder'' while maintaining a certain degree of naturalness.

Paper Structure

This paper contains 16 sections, 2 figures, 5 tables.

Table of Contents

  1. Introduction
  2. Datasets
  3. The URMP dataset
  4. The Bach10 dataset
  5. The proposed FR109 dataset
  6. Method
  7. Experiments
  8. Training
  9. Data preprocessing
  10. Evaluation
  11. Objective evaluation
  12. Subjective evaluation
  13. Analysis
  14. Mel spectrogram analysis
  15. Wiener entropy
  16. ...and 1 more sections

Figures (2)

  • Figure 1: Mel spectrograms of failed recorder music, the red rectangular parts show the inharmonic partials.
  • Figure 2: The Mel spectrograms of a failed-music style transfer example. (a) The Mel spectrogram of the source audio, which is performed by a saxophone; (b) The Mel spectrogram of the converted audio (to failed recorder) by StarGAN; (c) The Mel spectrogram of the converted audio (to failed recorder) by VAE-GAN. (d) The Mel spectrogram of the converted audio (to failed recorder) by DDSP.