The Nijmegen Corpus of Casual Spanish

Spanish is one of the best documented languages in the world. However, no large corpus of casual Spanish suitable for detailed phonetic analysis is available to our knowledge. The goal of this article is to introduce the Nijmegen Corpus of Casual Spanish (NCCSp from now on), a new corpus designed to fill this gap. The corpus was designed taking the Nijmegen Corpus Casual French as a model [Torreira et al., in press], which was also collected in our lab. The uniqueness of the NCCSp can be characterized as follows: • It contains around 30 hours of casual conversations among groups of friends. This makes it possible to study a wide range of phenomena characteristic of casual speech. • It contains speech from 52 native Madrid Spanish speakers sharing a similar educational background. • It contains large amounts of data for every speaker (around 90 minutes of recorded speech for every group of three speakers). This allows researchers to study withinspeaker variability. • It is orthographically annotated. • It contains video as well as audio data, which can be used by researchers interested in the use of facial and body gestures during verbal communication. The following sections provide a detailed description of the creation and transcription the NCCSp.