You are currently logged in as an
Institutional Subscriber.
If you would like to logout,
please click on the button below.
Home / Publications / E-library page
Only AES members and Institutional Journal Subscribers can download
Computer musicians refer to mesostructures as the intermediate levels of articulation between the microstructure of waveshapes and the macrostructure of musical forms. Examples of mesostructures include melody, arpeggios, syncopation, polyphonic grouping, and textural contrast. Despite their central role in musical expression, they have received limited attention in recent applications of deep learning to the analysis and synthesis of musical audio. Currently, autoencoders and neural audio synthesizers are only trained and evaluated at the scale of microstructure, i.e., local amplitude variations up to 100 ms or so. In this paper, the authors formulate and address the problem of mesostructural audio modeling via a composition of a differentiable arpeggiator and time-frequency scattering. The authors empirically demonstrate that time--frequency scattering serves as a differentiable model of similarity between synthesis parameters that govern mesostructure. By exposing the sensitivity of short-time spectral distances to time alignment, the authors motivate the need for a time-invariant and multiscale differentiable time--frequency model of similarity at the level of both local spectra and spectrotemporal modulations.
Author (s): Vahidi, Cyrus; Han, Han; Wang, Changhong; Lagrange, Mathieu; Fazekas, György; Lostanlen, Vincent
Affiliation:
"Centre for Digital Music, Queen Mary University of London, London, UK; Nantes Université, École Centrale Nantes, Centre National de la Recherche Scientifique (CNRS), Laboratoire desSciences du Numérique de Nantes (LS2N), UMR 6004, F-44000 Nantes, France; Nantes Université, École Centrale Nantes, Centre National de la Recherche Scientifique (CNRS), Laboratoire desSciences du Numérique de Nantes (LS2N), UMR 6004, F-44000 Nantes, France; Nantes Université, École Centrale Nantes, Centre National de la Recherche Scientifique (CNRS), Laboratoire desSciences du Numérique de Nantes (LS2N), UMR 6004, F-44000 Nantes, France; Centre for Digital Music, Queen Mary University of London, London, UK; Nantes Université, École Centrale Nantes, Centre National de la Recherche Scientifique (CNRS), Laboratoire des Sciences du Numérique de Nantes (LS2N), UMR 6004, F-44000 Nantes, France"
(See document for exact affiliation information.)
Publication Date:
2023-09-06
Import into BibTeX
Permalink: https://aes2.org/publications/elibrary-page/?id=22233
(767KB)
Click to purchase paper as a non-member or login as an AES member. If your company or school subscribes to the E-Library then switch to the institutional version. If you are not an AES member Join the AES. If you need to check your member status, login to the Member Portal.
Vahidi, Cyrus; Han, Han; Wang, Changhong; Lagrange, Mathieu; Fazekas, György; Lostanlen, Vincent; 2023; Mesostructures: Beyond Spectrogram Loss in Differentiable Time--Frequency Analysis [PDF]; "Centre for Digital Music, Queen Mary University of London, London, UK; Nantes Université, École Centrale Nantes, Centre National de la Recherche Scientifique (CNRS), Laboratoire desSciences du Numérique de Nantes (LS2N), UMR 6004, F-44000 Nantes, France; Nantes Université, École Centrale Nantes, Centre National de la Recherche Scientifique (CNRS), Laboratoire desSciences du Numérique de Nantes (LS2N), UMR 6004, F-44000 Nantes, France; Nantes Université, École Centrale Nantes, Centre National de la Recherche Scientifique (CNRS), Laboratoire desSciences du Numérique de Nantes (LS2N), UMR 6004, F-44000 Nantes, France; Centre for Digital Music, Queen Mary University of London, London, UK; Nantes Université, École Centrale Nantes, Centre National de la Recherche Scientifique (CNRS), Laboratoire des Sciences du Numérique de Nantes (LS2N), UMR 6004, F-44000 Nantes, France"; Paper ; Available from: https://aes2.org/publications/elibrary-page/?id=22233
Vahidi, Cyrus; Han, Han; Wang, Changhong; Lagrange, Mathieu; Fazekas, György; Lostanlen, Vincent; Mesostructures: Beyond Spectrogram Loss in Differentiable Time--Frequency Analysis [PDF]; "Centre for Digital Music, Queen Mary University of London, London, UK; Nantes Université, École Centrale Nantes, Centre National de la Recherche Scientifique (CNRS), Laboratoire desSciences du Numérique de Nantes (LS2N), UMR 6004, F-44000 Nantes, France; Nantes Université, École Centrale Nantes, Centre National de la Recherche Scientifique (CNRS), Laboratoire desSciences du Numérique de Nantes (LS2N), UMR 6004, F-44000 Nantes, France; Nantes Université, École Centrale Nantes, Centre National de la Recherche Scientifique (CNRS), Laboratoire desSciences du Numérique de Nantes (LS2N), UMR 6004, F-44000 Nantes, France; Centre for Digital Music, Queen Mary University of London, London, UK; Nantes Université, École Centrale Nantes, Centre National de la Recherche Scientifique (CNRS), Laboratoire des Sciences du Numérique de Nantes (LS2N), UMR 6004, F-44000 Nantes, France"; Paper ; 2023 Available: https://aes2.org/publications/elibrary-page/?id=22233
@article{vahidi2023mesostructures:,
author={vahidi cyrus and han han and wang changhong and lagrange mathieu and fazekas györgy and lostanlen vincent},
journal={journal of the audio engineering society},
title={mesostructures: beyond spectrogram loss in differentiable time--frequency analysis},
year={2023},
volume={71},
issue={9},
pages={577-585},
month={september},}
TY – paper
TI – Mesostructures: Beyond Spectrogram Loss in Differentiable Time--Frequency Analysis
SP – 577 EP – 585
AU – Vahidi, Cyrus
AU – Han, Han
AU – Wang, Changhong
AU – Lagrange, Mathieu
AU – Fazekas, György
AU – Lostanlen, Vincent
PY – 2023
JO – Journal of the Audio Engineering Society
VO – 71
IS – 9
Y1 – September 2023