提升音乐剧歌词翻译质量，兼顾可唱性和韵律要求。
Sing it, Narrate it: Quality Musical Lyrics Translation
- 构建专用数据集训练评分模型评估译文质量。
- 两阶段训练+过滤机制，平衡可唱性与翻译准确率。
- 推理时优化整首歌翻译，适合音乐剧本地化项目。
音乐剧歌词翻译面临独特挑战：需在保证高翻译质量的同时满足长度、押韵等可唱性要求。现有方法常过度侧重可唱性而牺牲译文质量，影响音乐剧表达效果。本文提出一种新方法，在保持关键可唱性特征的前提下显著提升翻译质量。首先，构建专用数据集用于训练自动评估翻译质量的奖励模型；其次，采用两阶段训练流程并引入过滤技术，同步优化可唱性与翻译准确性；最后，设计推理阶段的优化框架，实现整首歌曲的高质量翻译。大量实验（含自动评估与人工评测）表明，该方法在多个指标上优于基线模型，且各组件有效性得到验证。
原文摘要 · Abstract (English)
Translating lyrics for musicals presents unique challenges due to the need to ensure high translation quality while adhering to singability requirements such as length and rhyme. Existing song translation approaches often prioritize these singability constraints at the expense of translation quality, which is crucial for musicals. This paper aims to enhance translation quality while maintaining key singability features. Our method consists of three main components. First, we create a dataset to train reward models for the automatic evaluation of translation quality. Second, to enhance both singability and translation quality, we implement a two-stage training process with filtering techniques. Finally, we introduce an inference-time optimization framework for translating entire songs. Extensive experiments, including both automatic and human evaluations, demonstrate significant improvements over baseline methods and validate the effectiveness of each component in our approach.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。