Transformation of audio embeddings into interpretable, concept-based representations

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Zhang, Alice, Thomaz, Edison, Lu, Lie
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866909584758669312
author Zhang, Alice
Thomaz, Edison
Lu, Lie
author_facet Zhang, Alice
Thomaz, Edison
Lu, Lie
contents Advancements in audio neural networks have established state-of-the-art results on downstream audio tasks. However, the black-box structure of these models makes it difficult to interpret the information encoded in their internal audio representations. In this work, we explore the semantic interpretability of audio embeddings extracted from these neural networks by leveraging CLAP, a contrastive learning model that brings audio and text into a shared embedding space. We implement a post-hoc method to transform CLAP embeddings into concept-based, sparse representations with semantic interpretability. Qualitative and quantitative evaluations show that the concept-based representations outperform or match the performance of original audio embeddings on downstream tasks while providing interpretability. Additionally, we demonstrate that fine-tuning the concept-based representations can further improve their performance on downstream tasks. Lastly, we publish three audio-specific vocabularies for concept-based interpretability of audio embeddings.
format Preprint
id arxiv_https___arxiv_org_abs_2504_14076
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Transformation of audio embeddings into interpretable, concept-based representations
Zhang, Alice
Thomaz, Edison
Lu, Lie
Sound
Machine Learning
Audio and Speech Processing
Advancements in audio neural networks have established state-of-the-art results on downstream audio tasks. However, the black-box structure of these models makes it difficult to interpret the information encoded in their internal audio representations. In this work, we explore the semantic interpretability of audio embeddings extracted from these neural networks by leveraging CLAP, a contrastive learning model that brings audio and text into a shared embedding space. We implement a post-hoc method to transform CLAP embeddings into concept-based, sparse representations with semantic interpretability. Qualitative and quantitative evaluations show that the concept-based representations outperform or match the performance of original audio embeddings on downstream tasks while providing interpretability. Additionally, we demonstrate that fine-tuning the concept-based representations can further improve their performance on downstream tasks. Lastly, we publish three audio-specific vocabularies for concept-based interpretability of audio embeddings.
title Transformation of audio embeddings into interpretable, concept-based representations
topic Sound
Machine Learning
Audio and Speech Processing
url https://arxiv.org/abs/2504.14076