Sing it, Narrate it: Quality Musical Lyrics Translation
This work addresses the challenge of high-quality song translation for musical productions, though it appears incremental as it builds on existing approaches with new training and optimization techniques.
The paper tackled the problem of translating musical lyrics by enhancing translation quality while maintaining singability constraints like length and rhyme, achieving significant improvements over baseline methods as validated through automatic and human evaluations.
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.