A Study on Human-AI Collaboration Modes in Lyric Translation from a Translator-Centered Perspective

Abstract

With the rapid development of artificial intelligence technology, machine translation has become an important tool for translators to enhance productivity. However, lyric translation, which demands high levels of musicality, poetic sensibility, and cultural adaptability, represents a complex act of artistic recreation. At present, translators lack systematic methodological guidance for using AI in translation, making it difficult to effectively leverage artificial intelligence to improve translation quality. To address this, this study adopts a translator-oriented perspective to construct a human-computer collaborative workflow led by the translator, consisting of three stages: “pre-translation planning and prompting—critical screening during translation—post-translation precise editing and artistic refinement.” The paper emphasizes that translators should guide AI in generating draft translations through structured instructions while reinforcing their dual roles as translation project managers and artistic decision-makers during the interaction. By applying this model to the translation practice of the lyrics 《爱在》 its comprehensive effectiveness in coordinating rhythm, transforming imagery, and preserving poetic meaning was validated. The study concludes that the key to improving lyric translation quality in the era of artificial intelligence lies in the translator's mastery of the tool. It provides an actionable methodological framework for lyric translation and offers theoretical reference and practical insights for translation teaching and translator competence development.