Enregistré dans:
Détails bibliographiques
Auteurs principaux: Puhach, Dariia, Payberah, Amir H., Székely, Éva
Format: Preprint
Publié: 2025
Sujets:
Accès en ligne:https://arxiv.org/abs/2508.13603
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866912544011059200
author Puhach, Dariia
Payberah, Amir H.
Székely, Éva
author_facet Puhach, Dariia
Payberah, Amir H.
Székely, Éva
contents Similar to text-based Large Language Models (LLMs), Speech-LLMs exhibit emergent abilities and context awareness. However, whether these similarities extend to gender bias remains an open question. This study proposes a methodology leveraging speaker assignment as an analytic tool for bias investigation. Unlike text-based models, which encode gendered associations implicitly, Speech-LLMs must produce a gendered voice, making speaker selection an explicit bias cue. We evaluate Bark, a Text-to-Speech (TTS) model, analyzing its default speaker assignments for textual prompts. If Bark's speaker selection systematically aligns with gendered associations, it may reveal patterns in its training data or model design. To test this, we construct two datasets: (i) Professions, containing gender-stereotyped occupations, and (ii) Gender-Colored Words, featuring gendered connotations. While Bark does not exhibit systematic bias, it demonstrates gender awareness and has some gender inclinations.
format Preprint
id arxiv_https___arxiv_org_abs_2508_13603
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Who Gets the Mic? Investigating Gender Bias in the Speaker Assignment of a Speech-LLM
Puhach, Dariia
Payberah, Amir H.
Székely, Éva
Computation and Language
Artificial Intelligence
Similar to text-based Large Language Models (LLMs), Speech-LLMs exhibit emergent abilities and context awareness. However, whether these similarities extend to gender bias remains an open question. This study proposes a methodology leveraging speaker assignment as an analytic tool for bias investigation. Unlike text-based models, which encode gendered associations implicitly, Speech-LLMs must produce a gendered voice, making speaker selection an explicit bias cue. We evaluate Bark, a Text-to-Speech (TTS) model, analyzing its default speaker assignments for textual prompts. If Bark's speaker selection systematically aligns with gendered associations, it may reveal patterns in its training data or model design. To test this, we construct two datasets: (i) Professions, containing gender-stereotyped occupations, and (ii) Gender-Colored Words, featuring gendered connotations. While Bark does not exhibit systematic bias, it demonstrates gender awareness and has some gender inclinations.
title Who Gets the Mic? Investigating Gender Bias in the Speaker Assignment of a Speech-LLM
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2508.13603