Plurilingual teaching and learning of Romance languages exploits the similarities between these languages to teach them contrastively and to raise the language awareness of the learner. Several European projects have been devoted to plurilingual teaching and learning of Romance languages. The materials developed in these projects do not involve Natural Language Processing (NLP) capabilities and almost exclusively focus on receptive skills. The research goal of my Ph.D. dissertation was the design, development, implementation and evaluation of an interactive plurilingual ICALL (Intelligent Computer-Assisted Language Learning) software system (ESPRIT) for contrastive learning of French, Spanish and Italian aimed at advanced learners. I investigated how techniques from NLP enhance the plurilingual teaching and learning of these languages. The ESPRIT toolset comprises dictionary tools, a concordancer, an input analysis and feedback module, custom-made animated grammar presentations and an authoring tool for animated text. Dictionary tools provide useful information on unrestricted texts. The concordancer gives extensive information about how a language term is used in different contexts. The input analysis and feedback module dynamically provides precise feedback on restricted learner input up to paragraph level. Custom-made animated grammar presentations and learning materials created with the animation authoring tool visualise contrastive grammatical properties and processes. ESPRIT represents an interactive and flexible learning environment and is designed for autonomous learning. Formative and summative evaluation processes provided learner assessment data of different components of ESPRIT. A web-based databasedriven evaluation platform developed for ESPRIT can easily be adapted to other evaluation projects.