mlsanityData to 7 Oct 2026

Generative AI-based data augmentation for improved bioacoustic classification in noisy environments

arXiv 2412.01530 ↗First posted 2 Dec 2024cs.SD eess.AS stat.APcited by 1

On the map: Diffusion models → Audio, speech and music

Abstract (arXiv)

Obtaining data to train robust artificial intelligence (AI)-based models for species classification can be challenging, particularly for rare species. Data augmentation can boost classification accuracy by increasing the diversity of training data and is cheaper to obtain than expert-labelled data. However, many classic image-based augmentation techniques are not suitable for audio spectrograms. We investigate two generative AI models as data augmentation tools to synthesise spectrograms and supplement audio data: Auxiliary Classifier Generative Adversarial Networks (ACGAN) and Denoising Diffusion Probabilistic Models (DDPMs). The latter performed particularly well in terms of both realism of generated spectrograms and accuracy in a resulting classification task. Alongside these new approaches, we present a new audio data set of 640 hours of bird calls from wind farm sites in Ireland, approximately 800 samples of which have been labelled by experts. Wind farm data are particularly challenging for classification models given the background wind and turbine noise. Training an ensemble of classification models on real and synthetic data combined compared well with highly confident BirdNET predictions. Each classifier we used was improved by including synthetic data, and classification metrics generally improved in line with the amount of synthetic data added. Our approach can be used to augment acoustic signals for more species and other land-use types, and has the potential to bring about advances in our capacity to develop reliable AI-based detection of rare species. Our code is available at https://github.com/gibbona1/SpectrogramGenAI.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Audio, speech and music, one of 14 directions in Diffusion models. Builds on 33 of its 98 references that are in the map. First posted on arXiv 2 Dec 2024.

Builds on

33 of 98 references.

References · 33

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020Denoising Diffusion Probabilistic Modelscited by 20,851
  3. 2021High-Resolution Image Synthesis with Latent Diffusion Modelscited by 18,694
  4. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,267
  5. 2015Distilling the Knowledge in a Neural Networkcited by 10,076
  6. 2015Deep Unsupervised Learning using Nonequilibrium Thermodynamicscited by 7,005
  7. 2022Generative Adversarial Networkscited by 5,603
  8. 2018MobileNetV2: Inverted Residuals and Linear Bottleneckscited by 4,773
  9. 2017Neural Discrete Representation Learningcited by 4,650
  10. 2021Zero-Shot Text-to-Image Generationcited by 4,276
  11. 2016Improved Techniques for Training GANscited by 2,938
  12. 2020Knowledge Distillation: A Surveycited by 1,578
21 more
  1. 2019Semantic Image Synthesis with Spatially-Adaptive Normalizationcited by 1,295
  2. 2016Conditional Image Synthesis With Auxiliary Classifier GANscited by 562
  3. 2019MelGAN: Generative Adversarial Networks for Conditional Waveform Synthesiscited by 351
  4. 2018A Note on the Inception Scorecited by 225
  5. 2018Adversarial Audio Synthesiscited by 222
  6. 2023CADS: Unleashing the Diversity of Diffusion Models through Condition-Annealed Samplingcited by 124
  7. 2018Train Sparsely, Generate Densely: Memory-efficient Unsupervised Training of High-resolution Temporal GANcited by 120
  8. 2016Neural Autoregressive Distribution Estimationcited by 94
  9. 2022Adapting Pretrained Vision-Language Foundational Models to Medical Imaging Domainscited by 85
  10. 2021Computational bioacoustics with deep learning: a review and roadmapcited by 78
  11. 2023Global birdsong embeddings enable superior transfer learning for bioacoustic classificationcited by 51
  12. 2019Lower Dimensional Kernels for Video Discriminatorscited by 37
  13. 2020A Simple Fine-tuning Is All You Need: Towards Robust Deep Learning Via Adversarial Fine-tuningcited by 30
  14. 2021SpecMix : A Mixed Sample Data Augmentation method for Training withTime-Frequency Domain Featurescited by 30
  15. 2018Deep Convolutional Neural Network with Mixup for Environmental Sound Classificationcited by 15
  16. 2017Raw Waveform-based Audio Classification Using Sample-level CNN Architecturescited by 12
  17. 2023Self-Supervised Learning for Few-Shot Bird Sound Classificationcited by 11
  18. 2014Automatic large-scale classification of bird sounds is strongly improved by unsupervised feature learningcited by 4
  19. 2023Exploring The Potential Of GANs In Biological Sequence Analysiscited by 4
  20. 2024Mixture of Mixups for Multi-label Classification of Rare Anuran Soundscited by 4
  21. 2022NEAL: An open-source tool for audio annotationcited by 2

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026CoarseSoundNet: Building a reliable model for ecological soundscape analysiscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Data Augmentation for Environmental Sound Classification Using Diffusion Probabilistic Model with Top-k Selection Discriminator
  2. 2025Advancing Marine Bioacoustics with Deep Generative Models: A Hybrid Augmentation Strategy for Southern Resident Killer Whale Detection
  3. 2024A Benchmark of Generative Methods for Zero-Shot Environmental Sound Classification
  4. 2025Synthetic data enables context-aware bioacoustic sound event detection
  5. 2026Synthetic Data Augmentation for Medical Audio Classification: A Preliminary Evaluation
  6. 2026When Generative Augmentation Hurts: A Benchmark Study of GAN and Diffusion Models for Bias Correction in AI Classification Systems
  7. 2025State Space Models for Bioacoustics: A Comparative Evaluation with Transformers
  8. 2024Synthio: Augmenting Small-Scale Audio Classification Datasets with Synthetic Data