This paper is concerned with music enhancement by removal of coding artifacts and recovery of acoustic characteristics that preserve the sound quality of the original music content. In order to achieve this, we propose a novel convolution neural network (CNN) architecture called FTD (Frequency-Time Dependent) CNN, which utilizes correlation and context information across spectral and temporal dependency for music signals. Experimental results show that both subjective and objective sound quality metrics are significantly improved. This unique way of applying a CNN to exploit global dependency across frequency bins may effectively restore information that is corrupted by coding artifacts in compressed music content.

Authors: Porov, Anton; Oh, Eunmi; Choo, Kihyun; Sung, Hosang; Jeong, Jonghoon; Osipov, Konstantin; Francois, Holly
Affiliations: PDMI RAS, St. Petersburg, Russia; Samsung Electronics Co., Ltd., Seoul, Korea; Samsung Electronics R&D Institute UK, Staines-Upon Thames, Surrey, UK(See document for exact affiliation information.)
AES Convention: 145 (October 2018) Paper Number: 10036
Publication Date: October 7, 2018 Import into BibTeX
Subject: Signal Processing—Part 1
Permalink: https://www.aes.org/e-lib/browse.cfm?elib=19762

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 and would like to subscribe to the E-Library then Join the AES!

This paper costs $33 for non-members and is free for AES members and E-Library subscribers.

Learn more about the AES E-Library

E-Library Location:   /conv/145/10036.pdf

Start a discussion about this paper!