The Binding Effect: Analyzing How Multi-Dimensional Cues Form Gender Bias in Instruction TTS

TL;DR

This paper reveals complex gender biases in instruction-based TTS systems arising from multi-dimensional social cues, emphasizing the importance of compositional analysis over univariate testing to understand and mitigate bias.

Contribution

It introduces a multi-dimensional modeling approach for prompts, uncovering interaction effects and biases in open-source ITTS models that are overlooked by traditional univariate evaluations.

Findings

Bias patterns are influenced by social cue interactions.

Pre-trained text encoders' semantic priors contribute to biases.

Diversity prompts do not eliminate entrenched biases.

Abstract

Current bias evaluations in Instruction Text-to-Speech (ITTS) often rely on univariate testing, overlooking the compositional structure of social cues. In this work, we investigate gender bias by modeling prompts as combinations of Social Status, Career stereotypes, and Persona descriptors. Analyzing open-source ITTS models, we uncover systematic interaction effects where social dimensions modulate one another, creating complex bias patterns missed by univariate baselines. Crucially, our findings indicate that these biases extend beyond surface-level artifacts, demonstrating strong associations with the semantic priors of pre-trained text encoders and the skewed distributions inherent in training data. We further demonstrate that generic diversity prompting is insufficient to override these entrenched patterns, underscoring the need for compositional analysis to diagnose latent risks in generative speech.