Bridging Neural Representations and Symbolic Semantics in Audio Classification

Avishek Das, Pascal HitzlerarXiv 2026 · arXiv

Abstract

Deep audio classifiers can achieve strong predictive performance, but their hidden representations remain difficult to interpret. Existing audio explainability methods often focus on input-level attribution, such as waveform segments, spectrogram regions, or acoustic features, and provide limited insight into what hidden neurons encode. Interpreting hidden neurons at the concept level can provide a more meaningful view of what an audio model learns internally. This paper investigates whether hidden neurons in audio classifiers can be associated with human-understandable symbolic acoustic concepts. A model-agnostic neurosymbolic framework is presented for analyzing the final hidden layer of audio classifiers, containing 32 neurons, through symbolic concept induction. The framework is evaluated on two audio classification settings: music genre classification using GTZAN dataset and musical instrument classification using IRMAS dataset. Results show that several hidden neurons can be associated with meaningful induced concepts, and most confirmed neuron-concept pairs are statistically supported using the Mann-Whitney U test. These findings suggest that symbolic concept induction is a promising direction for explaining latent audio representations beyond input-level attribution.
BibTeX
@misc{das2026bridging,
  author = {Avishek Das and Pascal Hitzler},
  title = {{Bridging Neural Representations and Symbolic Semantics in Audio Classification}},
  year = {2026},
  eprint = {2610.40414},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2610.40414},
}

Loading related content...

Figure