Automatic Song Translation for Tonal Languages

被引:0
作者
Guo, Fenfei [1 ]
Zhang, Chen [2 ]
Zhang, Zhirui [3 ]
He, Qixin [4 ]
Zhang, Kejun [2 ]
Xie, Jun [5 ]
Boyd-Graber, Jordan [6 ]
机构
[1] Univ Maryland, College Pk, MD 20742 USA
[2] Zhejiang Univ, Hangzhou, Peoples R China
[3] Tencent AI Lab, Bellevue, WA USA
[4] Purdue Univ, W Lafayette, IN 47907 USA
[5] Alibaba DAMO Acad, Hangzhou, Peoples R China
[6] Univ Maryland, CS, iSch, UMIACS,LSC, College Pk, MD 20742 USA
来源
FINDINGS OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2022) | 2022年
关键词
D O I
暂无
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
This paper develops automatic song translation (AST) for tonal languages and addresses the unique challenge of aligning words' tones with melody of a song in addition to conveying the original meaning. We propose three criteria for effective AST-preserving meaning, singability and intelligibility-and design metrics for these criteria. We develop a new benchmark for English-Mandarin song translation and develop an unsupervised AST system, Guided AliGnment for Automatic Song Translation (GagaST), which combines pre-training with three decoding constraints. Both automatic and human evaluations show GagaST successfully balances semantics and singability.
引用
收藏
页码:729 / 743
页数:15
相关论文
共 55 条
[1]  
[Anonymous], 2016, ARXIV160406274
[2]  
[Anonymous], 2018, P NAACL 2018
[3]  
Bahdanau D, 2016, Arxiv, DOI arXiv:1409.0473
[4]  
Bellos David., 2013, Translation: Translators on Their Work and What It Means, P31
[5]  
Chen Yihao, 2020, 2020 IEEE INT S MULT
[6]  
Franzon Johan., 2008, TRANSLATOR, V14, P373, DOI DOI 10.1080/13556509.2008.10799263
[7]  
Genzel Dmitriy., 2010, Proceedings of the 2010 Conference on Empirical Methods in Natural Language Processing, P158
[8]  
Gervas P., 2000, Proceedings of the AISB-00 Symposium on Creative Cultural Aspects of AI, P93
[9]   Hafez: an Interactive Poetry Generation System [J].
Ghazvininejad, Marjan ;
Shi, Xing ;
Priyadarshi, Jay ;
Knight, Kevin .
PROCEEDINGS OF THE 55TH ANNUAL MEETING OF THE ASSOCIATION FOR COMPUTATIONAL LINGUISTICS (ACL 2017): SYSTEM DEMONSTRATIONS, 2017, :43-48
[10]  
Ghazvininejad Marjan., 2016, Proceedings of EMNLP, P1183, DOI 10.18653/v1/D16-1126