인간-AI 협업을 위한 화면해설 포스트에디팅 전략 연구 - 넷플릭스 「오징어 게임」 사례 분석을 중심으로

초록

This study explores post-editing strategies for human-AIcollaboration in audio description (AD) in the context of growingmedia accessibility demands and advances in generative AI. From amultimodal translation perspective, it analyzes the limitations ofAI-generated drafts and the role of human intervention inreconstructing narrative meaning. Using Squid Game and LeeSeo-jin's Dalla Dalla on Netflix as case studies, AI-generated scriptsproduced by Gemini 3.1 Pro were compared with official AD scripts. A three-dimensional framework was applied: accuracy of signs,narrative cohesion, and stylistic and audience appropriateness. Theresults show that “Remove” and “Shorten” are the most frequentstrategies used to control excessive description. The AI demonstrateslimitations in information prioritization, narrative contextmanagement, and cross-cut editing, leading to omission andover-description. Human editors reorganize narrative focus,reconnect sequences, and selectively transform on-screen text. These findings indicate that AD post-editing functions as a processof narrative and effective reconstruction rather than mere textualcorrection, highlighting the potential of a collaborative model thatcombines AI efficiency with human interpretive expertise.

키워드