NLP Tools

Tokeniser and POS Tagger for Maltese

This POS tagger uses an SVM with a linear kernel, trained over manually tagged text (ca. 93k tokens, 5200 sentences) from different genres.
The tagset is documented here.

Automatic grapheme-to-phoneme trasnscription

This is a rule-based G2P tool built as part of the MASRI Project.

Numbers to words

Conversion of numbers to their textual representation in Maltese. This tools was built as part of the MASRI Project.

Educational tools and fun stuff

Random Maltese Sentence Generator

Albert Gatt's Maltese random sentence generator: give the program information about words and get it to generate random sentences using a simple context-free grammar.

LingLab Language Guessing Game

Claudia Borg's LingLab language guessing game: listen to songs and guess the language they are in.