About this document

Localizing Audio Concepts in CNNs by Vinoo Alluri is a document available to read on EtoBox.

This paper investigates the interpretability of audio convolutional neural networks (CNNs), particularly focusing on the CLAP model, which processes audio data for tasks like keyword spotting and genre classification. The authors propose a framework to analyze neuron clusters within layers of the CNN to understand how different layers contribute to various levels of auditory abstraction, revealing insights into hierarchical encoding similar to that in human auditory processing. The study highlights the need

Author

Vinoo Alluri

Language

EN