The development of DOTI (Data of Oral Teletandem Interaction)

Teletandem1 (Telles and Vassallo, 2006) is a Voice Over Internet Protocol (VoIP) communicative activity in which two speakers are involved, each of whom is an expert in one language and who wishes to learn the language of the interlocutor. “Virtual meetings” which last one hour are organized weekly; students speak half of the time in their L1, and the other half in the L2. Teletandem is also a growing field of research, and the related data, collected by video-recording the conversations between two participants, are an interesting resource for analysing communication and learning processes. In order to build a databank (DOTI – Data of Oral Teletandem Interactions), we collected data from Sao Paolo State University at São José do Rio Preto (Brazil: languages Portuguese/English), and from the University of Salento (Italy: languages Italian/ English). DOTI is currently composed of about 700 hours of video data from, oral teletandem sessions. The current paper describes: i) the state of the art with regard to developing a databank with video recorded oral sessions, as well as chat conversations; ii) teletandem as an interaction space; iii) different learning scenarios and microtasks that might influence the type of data and, in turn, metadata, in this context.