Amphibian Sounds Generating Network Based on Adversarial Learning

Sangwook Park, Mounya Elhilali, David K. Han, Hanseok Ko

Research output: Contribution to journalArticlepeer-review

Abstract

This letter proposes a generative network based on adversarial learning for synthesizing short-time audio streams and investigates the effectiveness of data augmentation for amphibian call sounds classification. Based on Fourier analysis, the generator is designed by a multi-layer perceptron composed of frequency basis learning layers and an output layer, and a discriminator is constructed by a convolutional neural network. Additionally, regularization on weights is introduced to train the networks with practical data that includes some disturbances. Synthetic audio streams are evaluated by quantitative comparison using inception score, and classification results are compared for real versus synthetic data. In conclusion, the proposed generative network is shown to produce realistic sounds and therefore useful for data augmentation.

Original languageEnglish
Article number9072273
Pages (from-to)640-644
Number of pages5
JournalIEEE Signal Processing Letters
Volume27
DOIs
Publication statusPublished - 2020 Apr 20

Keywords

  • adversarial networks
  • audio stream generation
  • Generative model
  • Wasserstein distance

ASJC Scopus subject areas

  • Signal Processing
  • Electrical and Electronic Engineering
  • Applied Mathematics

Fingerprint Dive into the research topics of 'Amphibian Sounds Generating Network Based on Adversarial Learning'. Together they form a unique fingerprint.

Cite this