↻
Michael Saxon reposted
Vagrant Gautam
@dippedrusk.com
A raciolinguistic finding that has lived rent-free in my head for years is: People rate the exact same English audio as less intelligible when shown a brown person's face rather than a white face. In work led by Maya Nachesa, we see whether multimodal speech recognition models…
AI Weekly's analysis
→
- Swapping the face shown alongside identical audio shifts transcription accuracy by up to 4.05 word error rate points in mWhisper-Flamingo and Gemini.
- The gaps appear across self-declared gender, ethnicity, and their intersection, the authors report.
- The paper is described as the first bias evaluation of multimodal speech recognition.
Read full analysis →
View on Bluesky →