RedVox: Safety and Fairness Gaps in Speech Models Across Languages
For researchers and practitioners deploying speech models globally, this work highlights critical safety and fairness gaps in multilingual and naturalistic conditions.
The paper identifies that speech models are rarely evaluated for safety and fairness in non-English settings, with only 8% of model releases documenting multilingual analysis. It introduces RedVox, a multilingual benchmark covering five languages, and finds that vulnerabilities worsen in non-English languages and with spoken input across eight state-of-the-art models.
Speech-capable models are increasingly deployed in real-world applications across languages. Yet their safety and fairness beyond English settings and under naturalistic conditions remain understudied. We survey safety reporting practices across state-of-the-art speech model releases, finding that only 8% document any multilingual analysis. To address this gap, we introduce RedVox, a multilingual safety and fairness benchmark for audio and speech built on real voices, covering unsafe and unfair stereotypical requests across five languages (English, French, Italian, Spanish, and German). Evaluating eight state-of-the-art models, we find that vulnerabilities persist even under non-adversarial conditions, worsen in non-English languages, and are amplified when the request comes from a spoken input. Finally, by surveying the participants who contributed to RedVox, we document the unique personal and privacy challenges of collecting speech data with human participants, pointing to broader sociotechnical challenges in naturalistic speech safety research.