CREAM is an interdisciplinary project (linguistics and informatics) which aims at proposing new methods for language documentation, in particular for creole languages, using recent Machine Learning techniques. It aims at : - enriching existing corpora in different Creole languages (Kriol: Guinea-Bissau and Casamance, Jamaican Creole, French Creoles (Guadeloupe and Martinique)) with spoken in the respective lexifying languages (Portuguese, English French). - developping speech processing systems that allow (1) to search speech collections using spoken query examples (keyword spotting), (2) to perform bilingual alignement from spoken corpora (spoken alignment).
Monsieur Emmanuel Schang (Laboratoire Ligérien de Linguistique)
The author of this summary is the project coordinator, who is responsible for the content of this summary. The ANR declines any responsibility as for its contents.
LLL Laboratoire Ligérien de Linguistique
LIG Laboratoire d'Informatique de Grenoble
Help of the ANR 178,200 euros
Beginning and duration of the scientific project: - 48 Months