Bias in the Ear of the Listener: Assessing Sensitivity in Audio Language Models Across Linguistic, Demographic, and Positional Variations
Preprint
Abstract
This work presents the first systematic investigation of speech bias in multilingual MLLMs. We construct and release the BiasInEar dataset, a speech-augmented benchmark based on Global MMLU Lite, spanning English, Chinese, and Korean, balanced by gender and accent, and totaling 70.8 hours ($\approx$4,249 minutes) of speech with 11,200 questions. Using four complementary metrics (accuracy, entropy, APES, and Fleiss' $κ$), we evaluate nine representative models under linguistic (language and accent), demographic (gender), and structural (option order) perturbations. Our findings reveal that MLLMs are relatively robust to demographic factors but highly sensitive to language and option order, suggesting that speech can amplify existing structural biases. Moreover, architectural design and reasoning strategy substantially affect robustness across languages. Overall, this study establishes a unified framework for assessing fairness and robustness in speech-integrated LLMs, bridging the gap between text- and speech-based evaluation. The resources can be found at https://github.com/ntunlplab/BiasInEar.Keywords
All Related Versions
This publication has 20 references indexed in Scilit:
- ConversAR: Exploring Embodied LLM-Powered Group Conversations in Augmented Reality for Second Language Learners2025
- Evaluating OpenAI's Whisper ASR: Performance analysis across diverse accents and speaker traitsJASA Express Letters, 2024
- Uncovering Stereotypes in Large Language Models: A Task Complexity-based Approach2024
- Twists, Humps, and Pebbles: Multilingual Speech Recognition Models Exhibit Gender Performance Gaps2024
- Having Beer after Prayer? Measuring Cultural Bias in Large Language Models2024
- Unveiling Selection Biases: Exploring Order and Token Sensitivity in Large Language Models2024
- SpeechGPT: Empowering Large Language Models with Intrinsic Cross-Modal Conversational Abilities2023
- Racial disparities in automated speech recognitionProceedings of the National Academy of Sciences, 2020
- Measuring nominal scale agreement among many raters.Psychological Bulletin, 1971
- A Mathematical Theory of CommunicationBell System Technical Journal, 1948