Toward Fair Speech Technologies: A Comprehensive Survey of Bias and Fairness in Speech AI

TL;DR

This survey comprehensively reviews bias and fairness issues in speech AI, proposing a unified framework, formal definitions, evaluation metrics, and mitigation strategies tailored to speech-specific challenges.

Contribution

It introduces a formalized set of fairness definitions for speech, links evaluation metrics to these definitions, and systematically categorizes bias sources and mitigation methods in speech AI.

Findings

Identifies speech-specific bias mechanisms like channel bias and annotation subjectivity.

Provides a decision tree for selecting appropriate fairness evaluation metrics.

Maps mitigation strategies to specific bias sources in speech processing.

Abstract

Speech technologies are deployed in high-stakes settings, yet fairness concerns remain fragmented across tasks and disciplines. Existing surveys either adopt a general machine-learning perspective that overlooks speech-specific properties or focus on a single task, missing failure patterns shared across the speech domain. Synthesizing over 400 studies spanning generation and perception tasks and emerging speech-language models, this survey presents a unified framework that links formal fairness definitions to evaluation, diagnosis, and mitigation. We formalize seven fairness definitions adapted to the speech modality and organize the field's conceptual evolution through three paradigms: Robustness, Representation, and Governance. We then ground evaluation metrics in the mathematical cores of these definitions and offer a decision tree for metric selection. We diagnose bias sources along the speech processing pipeline, surfacing speech-specific mechanisms such as channel bias as a demographic proxy and annotation subjectivity in emotion labels. We systematize mitigation strategies across four intervention stages, mapping each to the diagnosed sources. Finally, we identify open challenges and propose directions for future research.