Language Resources for Spanish - Spanish Sign Language (LSE) translation

Abstract

This paper describes the development of a Spanish Spanish Sign Language (LSE) translation system. Firstly, it describes the first Spanish Spanish Sign Language (LSE) parallel corpus focused on two specific domains: the renewal of the Identity Document and Driver’s License. This corpus includes more than 4,000 Spanish sentences (in these domains), their LSE translation and a video for each LSE sentence with the sign language representation. This corpus also contains more than 700 sign descriptions in several sign writing specifications. The translation system developed with this corpus consists of two modules: a Spanish into LSE translation module that is composed of a speech recognizer (for decoding the spoken utterance into a word sequence), a natural language translator (for converting a word sequence into a sequence of signs) and a 3D avatar animation module (for playing back the signs). The second module is a Spanish generator from LSE made up of a visual interface (for specifying a sequence of signs in sign writing), a language translator (for generating the sequence of words in Spanish) and a text to speech converter. For each language translation, the system uses three technologies: an example based strategy, a rule based translation method and a statistical translator.