Software to support research and development of spoken dialogue systems

The development of a spoken dialogue system requires the integration of the various components of spoken language technology, such as speech recognition, natural language processing, dialogue modelling, and speech synthesis. Recently several toolkits have been developed that provide support for this process, enabling developers who have no specialist knowledge of the component technologies to produce working spoken dialogue systems with relative ease. This paper reports on the use of CSLU’s RAD (Rapid Application Developer) to provide practical experience for undergraduate students taking courses in spoken dialogue systems. Two groups of students were involved students of linguistics, speech and language therapy, and communication, on the one hand, and students of computational linguistics and computing science. The paper describes the use of the toolkit for students with these different degrees of competence in computing and reports on plans for future work with the toolkit.