Incremental generation for real-time applications

Abstract

The acceptance of natural language generation systems strongly depends on their capability to facilitate the exchange of information with human users. Current generation systems consider the influence of situational factors on the content and the form of the resulting utterances. However, the need to time their processing flexibly is usually neglected although temporal factors play a central part when directly addressing a human communication partner. A short response time of a system is crucial for its effective use. Furthermore, some applications - e.g., the simultaneous description of ongoing events - even necessitate the interleaving of input consumption and output production, i.e. the use of an incremental processing mode. We claim that incremental processing is a central design principle for developing flexible and efficient generators for speech output. We discuss the advantages of parallel processing for incremental generation and several aspects of control of the generator. An extension of Tree Adjoining Grammar is introduced as an adequate representation formalism for incremental syntactic generation. We present the system VM-GEN - an incremental and parallel syntactic generator based on Tree Adjoining Grammars. It offers flexible input and output interfaces that are adaptable to the requirements of the surrounding system by coping with varying sizes of input and output increments. The system's ability to produce fluent speech is a step towards approximating human language performance.