A Hierarchical Attention Based Seq2Seq Model for Chinese Lyrics Generation

被引：12
作者：
Fan, Haoshen [1 ,2 ]
Wang, Jie [2 ]
Zhuang, Bojin [2 ]
Wang, Shaojun [2 ]
Xiao, Jing [2 ]
机构：
[1] Univ Sci & Technol China, Hefei, Peoples R China
[2] Ping An Technol Shenzhen Co Ltd, Shenzhen, Peoples R China
来源：
PRICAI 2019: TRENDS IN ARTIFICIAL INTELLIGENCE, PT III | 2019年 / 11672卷
关键词：
Natural language generation; Seq2Seq; Gate recurrent unit; Attention;
D O I：
10.1007/978-3-030-29894-4_23
中图分类号：
TP18 [人工智能理论];
学科分类号：
081104 ; 0812 ; 0835 ; 1405 ;
摘要：
In this paper, we comprehensively study on context-aware generation of Chinese song lyrics. Conventional text generative models generate a sequence or sentence word by word, failing to consider the contextual relationship between sentences. Taking account into the characteristics of lyrics, a hierarchical attention based Seq2Seq (Sequence-to-Sequence) model is proposed for Chinese lyrics generation. With encoding of word-level and sentence-level contextual information, this model promotes the topic relevance and consistency of generation. A large Chinese lyrics corpus is also leveraged for model training. Eventually, results of automatic and human evaluations demonstrate that our model is able to compose complete Chinese lyrics with one united topic constraint.
引用
收藏
页码：279 / 288
页数：10
相关论文
共 21 条
  • [1] [Anonymous], 2016, P 26 INT C COMP LING
  • [2] [Anonymous], 2015, ICLR
  • [3] [Anonymous], 2014, COMPUTER SCI
  • [4] [Anonymous], 2016, C COMP NAT LANG LEAR
  • [5] [Anonymous], 2010, INTERSPEECH, DOI DOI 10.1016/J.CSL.2010.08.008
  • [6] [Anonymous], 2015, 3 INT C LEARN REPR I
  • [7] [Anonymous], 2015, P 2015 C EMP METH NA
  • [8] A neural probabilistic language model
    Bengio, Y
    Ducharme, R
    Vincent, P
    Jauvin, C
    [J]. JOURNAL OF MACHINE LEARNING RESEARCH, 2003, 3 (06) : 1137 - 1155
  • [9] Natural language processing
    Chowdhury, GG
    [J]. ANNUAL REVIEW OF INFORMATION SCIENCE AND TECHNOLOGY, 2003, 37 : 51 - 89
  • [10] Feng XC, 2018, PROCEEDINGS OF THE TWENTY-SEVENTH INTERNATIONAL JOINT CONFERENCE ON ARTIFICIAL INTELLIGENCE, P4078
123