A human-computer dialogue system for Māori language learning

In this paper we describe Te Kaitito, a bilingual human-machine dialogue system which supports conversational interactions in English and in Māori, the indigenous language of New Zealand. We concentrate in particular on an application of Te Kaitito in the field of computer-aided second language learning. The development of this tool has involved a group of researchers with very different backgrounds. Alistair Knott is a computational linguist; John Moorfield is an expert on the Māori language and on Māori language teaching; Tamsin Meaney is a lecturer in applied linguistics who has interests in computer aided language learning and indigenous languages and Lee-Luan Ng is currently conducting research for her PhD on how the cognitive styles of learners affect interactions with multi-media language learning courseware. We begin in Section 2 with a overview of the motivation behind the Te Kaitito system, and its intended role in the continued development of te reo Māori as a working language in New Zealand. In Section 3, we briefly survey the existing applications of natural language processing (NLP) in computer-aided language learning (CALL). In Section 4, we outline the architecture of Te Kaitito, and give some examples of interactions with the system. In Section 5, we describe some proposed extensions of the system to deal with language-learning dialogues, and some initial implementations of these ideas.