Présentation d’une démarche de valorisation et d’exploitation informatiques des corpus oraux. Le cas du logiciel Elan

Résumé

Agzul:

Di tezwart unadi-ya, yella-d wawal ɣef unbeddel ameqqran i d-yellan deg wallalen usekker asenselkam n wammuden i isemtawiyen ger imesli d tira n wammud. Deg weḥric-is amezwaru, nebder-d kra n tenfawin yesɛa Elan, diɣ neglem-d kra deg ugrudem-is, i nwala ɣur-s azal d ameqqran. Deg weḥric-is wis-sin, neṭṭef, d amedya, amek ara neg asnimek alɣasnay s Elan. Syin ɣer-s, nutlay-d ɣef tenfawin nniḍen, yellan deg Elan-CorpA, akka am usɣiwel n usnimek s umawal i d-yettakk a t-nefren. Di tagara-s, nesken-d kra iberdan unadi ɣef wayen yellan d isefka deg wammud i nesnimek di Elan d lexṣaṣ yellan deg-s, am uɣlluy ur d-ɣellint ara swa swa tilas imesli n wawalen d tira-nsen.

Abstract:

After the introduction, which emphasizes the upheaval that linguistics witnesses in data processing techniques corpus, which allow time synchronization between sound and annotations, the first part of this paper outlines the essential features that Elan is equipped and a brief presentation of the interface of this software. The second part, as an illustration, first shows how to use Elan for morphosyntactic annotation. Then it sets the additional features of ELAN-CORPA that simplify the interlinear in Elan for integrating a lexicon. Finally, it shows some options in Elan and discusses the limitations of the software, including the fact that it does not allow a precise location of the border segment of the sound signal on the note to which it refers that segment.