VoiceBBQ: Investigating Effect of Content and Acoustics in Social Bias of Spoken Language Model
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866911176146812928 |
|---|---|
| author | Choi, Junhyuk Oh, Ro-hoon Seol, Jihwan Kim, Bugeun |
| author_facet | Choi, Junhyuk Oh, Ro-hoon Seol, Jihwan Kim, Bugeun |
| contents | 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, 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 resists acoustic bias while 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_21108 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | VoiceBBQ: Investigating Effect of Content and Acoustics in Social Bias of Spoken Language Model Choi, Junhyuk Oh, Ro-hoon Seol, Jihwan Kim, Bugeun Computation and Language 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, 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 resists acoustic bias while 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. |
| title | VoiceBBQ: Investigating Effect of Content and Acoustics in Social Bias of Spoken Language Model |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2509.21108 |