The ECOLANG Multimodal Corpus enables a new approach to the study of language in its core ecological niche: face-to-face, multimodal environments.
Browse the ECOLANG Corpus Project Collection
The ECOLANG corpus provides audiovisual recordings and annotation of multimodal communicative behaviours by English-speaking adults engaged in semi-naturalistic conversation. One part of the corpus focuses on child-directed communication (n = 38 dyads), with caregivers interacting with their 3- to 4-year-old child. The other part focuses on adult-directed communication (n = 31 dyads), where an adult (designated speaker) interacts with another familiar adult (listener). In both parts, dyads talked about objects. The objects were either familiar or unfamiliar for the child/listener and taken from four categories: tools, animals, musical instruments and food (3 familiar and 3 unfamiliar from each category). Familiarity for children was established based on caregiver report, for adult listeners this was based on norms. Dyads discussed objects from each category twice for about 3-5 minutes; once while objects were physically present (e.g., the objects themselves or replicas), once whilst they were absent. Thus there were eight sessions in each recording (4 object categories x 2 states of object presence) lasting around 30-40 minutes. Caregiver/speaker behaviours are annotated, with all speech transcribed, gestures (representational gestures, pointing, object manipulations, pragmatic gestures and beat gestures) and gaze to objects coded.To obtain permission to view or download the video data (either viewing in your browser or downloading to your computer), you need to apply for a user licence. Futher information can be found on the ECOLANG Corpus web page.
Close
