Lost in Transcription: Subtitle Errors in Automatic Speech Recognition Reduce Speaker and Content Evaluations

2026/03/16 by Kowe Kadoma, Priyal Shrivastava, Mor Naaman · 1 voice
Computer Science · #cs.HC

paper · pdf · doi:10.1145/3772318.3790911

Abstract

Researchers have demonstrated that Automatic Speech Recognition (ASR) systems perform differently across demographic groups. In this work, we examined how subtitle errors affect evaluations of speakers and their content using a preregistered online experiment (N=207, U.S.-based crowdworkers). Participants watched speakers with various accents deliver a talk in which the subtitles were accurate or error-prone. Our results indicate that error-prone subtitles consistently reduce both speaker and content evaluations for all speakers. We did not see disparate impact between the accent groups, controlling for subtitle quality. Taken together, though, the findings of this short paper imply that speakers with accents for which ASR systems perform poorly are likely to be further penalized by viewers with lower evaluations.

Citations

• A Multi-Dialectal Dataset for German Dialect ASR and Dialect-to-Standard Speech Translation
• "It's not a representation of me": Examining Accent Bias and Digital Exclusion in Synthetic AI Voice Services
• Designing for Harm Reduction: Communication Repair for Multicultural Users' Voice Interactions
• Does Foreign-Accented Speech Affect Credibility? Evidence from the Illusory-Truth Paradigm
• Pre-trained Speech Processing Models Contain Human-Like Biases that Propagate to Speech Emotion Recognition
• Can Voice Assistants Be Microaggressors? Cross-Race Psychological Responses to Failures of Automatic Speech Recognition
• Sociotechnical Harms of Algorithmic Systems: Scoping a Taxonomy for Harm Reduction
• SubER: A Metric for Automatic Evaluation of Subtitle Quality
• Racial disparities in automated speech recognition
• End-to-end ASR: from Supervised to Semi-Supervised Learning with Modern Architectures
• A Framework for Understanding Sources of Harm throughout the Machine Learning Life Cycle
• Bias in computer systems
• Dual coding theory and education

Discussions

• In what seems to be the second "Lost In..." paper for your list 😮, @kkado.bsky.social just posted this one from us: arxiv.org/abs/2603.15807 Thanks for collecting these papers! [bsky, 2 points, 0 comments]

Related

• Lost in Transcription: Identifying and Quantifying the Accuracy Biases of Automatic Speech Recognition Systems Against Disfluent Speech
• Refining Transcripts With TV Subtitles by Prompt-Based Weakly Supervised Training of ASR
• Leveraging Broadcast Media Subtitle Transcripts for Automatic Speech Recognition and Subtitling
• Segmenting Subtitles for Correcting ASR Segmentation Errors
• You don't understand me!: Comparing ASR results for L1 and L2 speakers of Swedish