The Vinci Project is devoted to the generation of natural language and related phenomena. As well as developing the
ivi/vinci natural language generation system, we have undertaken generation-based studies of various linguistic phenomena, including second language learning, word formation, syntactic modelling, linguistic humour, semantic expressions and narrative generation. For details, see the Research overview and publications link below.
ivi/vinci also underlies the VinciLingua web-based language learning environment.
ivi/vinci software provides linguists with a collection of linguist-friendly metalanguages allowing them to model natural language phenomena and to generate utterances based on their models. Aspects of the metalanguages deal with semantics, syntax and morphology (both inflectional and derivational), as well as lexical relations. For the more technically minded, the system takes the form of a Chomsky Type 2 transformational grammar (which gives it more power than, for example, Finite State Machines), augmented by, among other things, attributes and conditions, lexical preselections and lexical pointers.
ivi/vinci is written in C and has been compiled for a variety of operating systems. We offer it without charge to any researchers who may want to make use of it. Information on how it works and how to use it is provided under the System documentation link below, and downloadable executable code may be obtained under the Downloads link below.
The project has been ongoing since the mid-1980's, led by Michael Levison (who unfortunately passed away in 2019) and Greg Lessard, who may be contacted at
vincilingua.ca. Over its history, it has involved a variety of participants in Computing Science and Linguistics. For a fuller history and details on participants, see the ivi and Vinci history and the Project members links below.
Those pressed for time should a) download the software for their operating system, b) read the basic description of the ivi editor and then c) read the gentle introduction to the Vinci generation commands. Those with more leisure should follow the links below.