Audio Inpainting of Music by Means of Neural Networks

Marafioti A, Holighaus N, Majdak P, Perraudin N. Audio Inpainting of Music by Means of Neural Networks. In: AES Convention 146. Audio Engineering Society; 2019. Paper 10170. Available from: https://aes.org/publications/elibrary-page/?id=20303

@inproceedings{Marafioti2019_20303,
  author = {Marafioti, Andrés and Holighaus, Nicki and Majdak, Piotr and Perraudin, Nathanaël},
  title = {{Audio Inpainting of Music by Means of Neural Networks}},
  booktitle = {AES Convention 146},
  note = {Paper 10170},
  year = {2019},
  month = mar,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=20303}
}

TY  - CPAPER
TI  - Audio Inpainting of Music by Means of Neural Networks
AU  - Marafioti, Andrés
AU  - Holighaus, Nicki
AU  - Majdak, Piotr
AU  - Perraudin, Nathanaël
T2  - AES Convention 146
M1  - Paper 10170
PY  - 2019
DA  - 2019/03/06
UR  - https://aes.org/publications/elibrary-page/?id=20303
PB  - Audio Engineering Society
LA  - en
AB  - We studied the ability of deep neural networks (DNNs) to restore missing audio content based on its context, a process usually referred to as audio inpainting. We focused on gaps in the range of tens of milliseconds. The proposed DNN structure was trained on audio signals containing music and musical instruments, separately, with 64-ms long gaps and represented by time-frequency (TF) coefficients. For music, our DNN significantly outperformed the reference method based on linear predictive coding (LPC), demonstrating a generally good usability of the proposed DNN structure for inpainting complex audio signals like music.
ER  -