AI speech assessors favor US and UK accents, UW study finds
A University of Washington study reported by TechNews found that four AI models recognized speech across American, British, Chinese, Indian, and Nigerian accents but scored American and British speakers most highly. The bias affected both pronunciation and content scores—even when instructed to score content alone—and often appeared in descriptions labeling US and UK speech clear or excellent while Chinese and Nigerian speech was unclear or difficult to understand.
The findings raise fairness concerns for AI-assisted hiring, education, presentations, interviews, and English-language assessments, and support adding accent-specific fairness testing to model evaluation.


華盛頓大學(University of Washington)最新研究指出,當人工智慧(AI)模型被用於面試、簡報或英語口說測驗評分時,說話者的「口音」恐將左右最終成績。研究團隊測試多款先進的語音與語言模型後發現,儘管這些系統都能準確辨識不同口音的說話內容,但在給分上卻明顯偏好美式與英式英語,而中文與奈及利亞口音則普遍拿下最低分。
這項研究共比較了五種口音:美式、英式、中式、印度與奈及利亞口音。測試結果顯示,無論是 AI 生成的合成語音,還是由真人朗讀相同的講稿,模型對美英口音的評價始終最高,印度口音居中,中文與奈及利亞口音的得分則敬陪末座。研究人員特別指出,這種落差不僅出現在「發音表現」的評分上,甚至連「內容分數」也連帶受到影響──即便研究人員已明確要求系統「只能根據說話內容來評分」。
#University of Washington#Gemini 2.5 Pro#Gemini 2.5 Flash#GPT-Audio-1.5#Qwen3-Omni