Document Type

Conference Paper


Available under a Creative Commons Attribution Non-Commercial Share Alike 4.0 International Licence



Publication Details

Proceedings of the TISLID 2010 Conference, Madrid, Spain October 2010.


Speech as an LSP: Many dialogues presented to language learners could be better described as ‘interleaved mini-monologues’, their purpose being to provide examples of grammatical sentences in realistic settings. Real dialogues, on the other hand, are worked out ‘live’, with neither speaker knowing in detail where the conversation will lead. Speaker interaction is marked to a large extent by prosody and often even good communicators sound disfluent if their half of the dialogue is judged in isolation. Dialogic fluency: The objective of dialoguing L1 speakers, however, is to realise a social or personal goal, with language only part of effective communication. Possibly the bulk of the communication devolves to prosody, shared knowledge and body language. Whereas this might not be a mainstream production goal for language learners, all users of English as an international language likely to come into contact with native speakers should be sensitised to native-speaker prosody. Influence of live dialogue on speech production: Given that the aim of an L1-L1 dialogue is not to provide learners with sample sentences, but rather to use language as a key factor in a social encounter, learners need a tool which will allow them to study the interaction of real dialogues. Of particular interest is the turn-taking behaviour of speakers, which is often flagged prosodically and produces utterances which, on the surface seem disfluent, but which on further analysis are seen to have an interactive function. The production of such a tool is the aim of the Dynamic Speech Corpus (DSC).


EU Lifelong Learning Programme