An Evolutionary-based Generative Approach for Audio Data Augmentation

Silvan Mertes, Alice Baird, Dominik Schiller, Bjorn W. Schuller, Elisabeth Andre

Publikation: Beitrag in Buch/Bericht/KonferenzbandKonferenzbeitragBegutachtung

19 Zitate (Scopus)

Abstract

In this paper, we introduce a novel framework to augment raw audio data for machine learning classification tasks. For the first part of our framework, we employ a generative adversarial network (GAN) to create new variants of the audio samples that are already existing in our source dataset for the classification task. In the second step, we then utilize an evolutionary algorithm to search the input domain space of the previously trained GAN, with respect to predefined characteristics of the generated audio. This way we are able to generate audio in a controlled manner that contributes to an improvement in classification performance of the original task. To validate our approach, we chose to test it on the task of soundscape classification. We show that our approach leads to a substantial improvement in classification results when compared to a training routine without data augmentation and training with uncontrolled data augmentation with GANs.

OriginalspracheEnglisch
TitelIEEE 22nd International Workshop on Multimedia Signal Processing, MMSP 2020
Herausgeber (Verlag)Institute of Electrical and Electronics Engineers Inc.
ISBN (elektronisch)9781728193205
DOIs
PublikationsstatusVeröffentlicht - 21 Sept. 2020
Extern publiziertJa
Veranstaltung22nd IEEE International Workshop on Multimedia Signal Processing, MMSP 2020 - Virtual, Tampere, Finnland
Dauer: 21 Sept. 202024 Sept. 2020

Publikationsreihe

NameIEEE 22nd International Workshop on Multimedia Signal Processing, MMSP 2020

Konferenz

Konferenz22nd IEEE International Workshop on Multimedia Signal Processing, MMSP 2020
Land/GebietFinnland
OrtVirtual, Tampere
Zeitraum21/09/2024/09/20

Fingerprint

Untersuchen Sie die Forschungsthemen von „An Evolutionary-based Generative Approach for Audio Data Augmentation“. Zusammen bilden sie einen einzigartigen Fingerprint.

Dieses zitieren