A Hierarchical Attention Based Seq2seq Model for Chinese Lyrics Generation
This addresses the challenge of context-aware text generation for Chinese lyrics, which is a domain-specific incremental improvement over conventional generative models.
The paper tackles the problem of generating Chinese song lyrics with consistent topics by proposing a hierarchical attention-based Seq2Seq model that encodes word-level and sentence-level context. Results from automatic and human evaluations show the model can compose complete Chinese lyrics with unified topic constraints.
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.