Natural language generation (NLG) is a critical component of spoken dialogueand it has a significant impact both on usability and perceived quality. MostNLG systems in common use employ rules and heuristics and tend to generaterigid and stylised responses without the natural variation of human language.They are also not easily scaled to systems covering multiple domains andlanguages. This paper presents a statistical language generator based on asemantically controlled Long Short-term Memory (LSTM) structure. The LSTMgenerator can learn from unaligned data by jointly optimising sentence planningand surface realisation using a simple cross entropy training criterion, andlanguage variation can be easily achieved by sampling from output candidates.With fewer heuristics, an objective evaluation in two differing test domainsshowed the proposed method improved performance compared to previous methods.Human judges scored the LSTM system higher on informativeness and naturalnessand overall preferred it to the other systems.
展开▼
