Sounds of the deep: How input representation, model choice, and dataset size influence underwater sound classification performance

  • Abdullah Olcay
  • , Paul R. White
  • , Jonathan M. Bull
  • , Denise Risch
  • , Benedict Dell
  • , Ellen L. White

Research output: Contribution to journalArticlepeer-review

Original languageEnglish
Pages (from-to)3017-3032
Number of pages15
JournalThe Journal of the Acoustical Society of America
Volume157
Issue number4
DOIs
Publication statusPublished - 1 Apr 2025

Keywords

  • deep neural networks
  • channel energy
  • convolutional neural network
  • data set size
  • energy normalization
  • model choice
  • representation model
  • sound classification
  • spectrograms
  • training data
  • underwater sound
  • article
  • background noise
  • classification
  • classifier
  • click
  • controlled study
  • diagnosis
  • human
  • male
  • marine environment
  • pharmaceutics
  • residual neural network
  • Scotland
  • sound
  • acoustic generators

Cite this