WavShape: Information-Theoretic Speech Representation Learning for Fair and Privacy-Aware Audio Processing

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Baser, Oguzhan, Tanriverdi, Ahmet Ege, Kale, Kaan, Chinchali, Sandeep P., Vishwanath, Sriram
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866915537777328128
author Baser, Oguzhan
Tanriverdi, Ahmet Ege
Kale, Kaan
Chinchali, Sandeep P.
Vishwanath, Sriram
author_facet Baser, Oguzhan
Tanriverdi, Ahmet Ege
Kale, Kaan
Chinchali, Sandeep P.
Vishwanath, Sriram
contents Speech embeddings often retain sensitive attributes such as speaker identity, accent, or demographic information, posing risks in biased model training and privacy leakage. We propose WavShape, an information-theoretic speech representation learning framework that optimizes embeddings for fairness and privacy while preserving task-relevant information. We leverage mutual information (MI) estimation using the Donsker-Varadhan formulation to guide an MI-based encoder that systematically filters sensitive attributes while maintaining speech content essential for downstream tasks. Experimental results on three known datasets show that WavShape reduces MI between embeddings and sensitive attributes by up to 81% while retaining 97% of task-relevant information. By integrating information theory with self-supervised speech models, this work advances the development of fair, privacy-aware, and resource-efficient speech systems.
format Preprint
id arxiv_https___arxiv_org_abs_2506_22789
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle WavShape: Information-Theoretic Speech Representation Learning for Fair and Privacy-Aware Audio Processing
Baser, Oguzhan
Tanriverdi, Ahmet Ege
Kale, Kaan
Chinchali, Sandeep P.
Vishwanath, Sriram
Sound
Artificial Intelligence
Audio and Speech Processing
Speech embeddings often retain sensitive attributes such as speaker identity, accent, or demographic information, posing risks in biased model training and privacy leakage. We propose WavShape, an information-theoretic speech representation learning framework that optimizes embeddings for fairness and privacy while preserving task-relevant information. We leverage mutual information (MI) estimation using the Donsker-Varadhan formulation to guide an MI-based encoder that systematically filters sensitive attributes while maintaining speech content essential for downstream tasks. Experimental results on three known datasets show that WavShape reduces MI between embeddings and sensitive attributes by up to 81% while retaining 97% of task-relevant information. By integrating information theory with self-supervised speech models, this work advances the development of fair, privacy-aware, and resource-efficient speech systems.
title WavShape: Information-Theoretic Speech Representation Learning for Fair and Privacy-Aware Audio Processing
topic Sound
Artificial Intelligence
Audio and Speech Processing
url https://arxiv.org/abs/2506.22789