@inproceedings{choi-etal-2025-voicebbq,
    title = "{V}oice{BBQ}: Investigating Effect of Content and Acoustics in Social Bias of Spoken Language Model",
    author = "Choi, Junhyuk  and
      Oh, Ro-hoon  and
      Seol, Jihwan  and
      Kim, Bugeun",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/author-page-yu-wang-polytechnic/2025.emnlp-main.1461/",
    doi = "10.18653/v1/2025.emnlp-main.1461",
    pages = "28713--28724",
    ISBN = "979-8-89176-332-6",
    abstract = "We introduce VoiceBBQ, a spoken extension of the BBQ (Bias Benchmark for Question answering) - a dataset that measures social bias by presenting ambiguous or disambiguated contexts followed by questions that may elicit stereotypical responses. Due to the nature of speech modality, social bias in Spoken Language Models (SLMs) can emerge from two distinct sources: 1) content aspect and 2) acoustic aspect. The dataset converts every BBQ context into controlled voice conditions, enabling per-axis accuracy, bias, and consistency scores that remain comparable to the original text benchmark. Using VoiceBBQ, we evaluate two SLMs{---}LLaMA-Omni and Qwen2-Audio{---}and observe architectural contrasts: LLaMA-Omni retains strong acoustic sensitivity, amplifying gender and accent bias, whereas Qwen2-Audio substantially dampens these cues while preserving content fidelity. VoiceBBQ thus provides a compact, drop-in testbed for jointly diagnosing content and acoustic bias across spoken language models."
}