(en) This paper comes within the scope of the Computer Assisted Language Learning framework, and addresses more especially the automation of dictation exercises. It presents a correction method of learners’ copies that is original in two ways. First, the method exploits the composition of finite-state automata, to both detect and analyze the errors. Second, it relies on an automatic morphosyntactic analysis of the original dictation, which makes it easier to produce diagnoses.
Beaufort, R., & Roekhaut, S. (2011). Le TAL au service de l’ALAO/ELAO. L’exemple des exercices de dictée automatisés. Actes de la 18me Conférence sur le Traitement Automatique des Langues Naturelles 2011. Published. TALN, Montpellier. https://hdl.handle.net/2078.5/162264