Speaking involves multitasking. Speakers must pace the type and amount of information given to their listeners, while putting together grammatical constructions and maintaining engagement with interlocutors. They achieve these goals in milliseconds, distilling layers of linguistic and sociocultural knowledge into intonation units (IUs) – chunks of speech identified by specific acoustic cues. The main objective of SPEAK (/spi:k/ ‘Convergence of syntactic, prosodic, and interactional units in conversation’) is to identify the basic units of organization of spoken discourse. To achieve this goal, the project combines corpus analysis and experimental methods to investigate to what degree intonation units correspond to units of syntax and communicative functions in spoken interaction.
Go to pageWe welcome collaborations from students interested in gaining hands-on research in linguistic data processing and linguistic analysis.
A new publication using conversational data in Kazakh language is now available
A publication featuring the description of the first module of the MCSKL corpus is now available in open access.
The first module of the Multimedia Corpus of Spoken Kazakh Language is now available for download and analysis.
With the support of the Marie Skłodowska-Curie Actions programme, SPEAK aims to uncover how intonation units, segments of speech that organize spoken discourse, connect to syntax and communication.
The talk discusses the creation of spoken corpora and several phenomena, including Russian/Sakha code-switching, word order in noun phrases, and preposition dropping in Russian spoken in Daghestan.