Listen and Speak Fairly: A Study on Semantic Gender Bias in Speech Integrated Large Language Models
This work addresses fairness issues in SILLMs for marginalized groups, but it is incremental as it focuses on evaluation rather than proposing new mitigation methods.
The study tackled gender bias in Speech Integrated Large Language Models (SILLMs) by evaluating it across four semantic tasks, revealing that bias levels are language-dependent and vary with evaluation methods, emphasizing the need for multiple approaches to assess biases comprehensively.
Speech Integrated Large Language Models (SILLMs) combine large language models with speech perception to perform diverse tasks, such as emotion recognition to speaker verification, demonstrating universal audio understanding capability. However, these models may amplify biases present in training data, potentially leading to biased access to information for marginalized groups. This work introduces a curated spoken bias evaluation toolkit and corresponding dataset. We evaluate gender bias in SILLMs across four semantic-related tasks: speech-to-text translation (STT), spoken coreference resolution (SCR), spoken sentence continuation (SSC), and spoken question answering (SQA). Our analysis reveals that bias levels are language-dependent and vary with different evaluation methods. Our findings emphasize the necessity of employing multiple approaches to comprehensively assess biases in SILLMs, providing insights for developing fairer SILLM systems.