The social inclusion of migrants is strongly dependent on the development of their skills in the language of the country of arrival. The CALL-IT project intends to promote the teaching and learning of Italian L2 by creating, testing and deploying digitally-based conversational applications, namely chatbots, which will run on mobile devices commonly used among foreigners who come to our country. This technological solution, along with the methodology devised for its employment, are intended to support seamless learning both inside and outside the classroom, contextualizing language learning in everyday situations experienced by migrants.


  • Stimulate use of the Italian language, inside and outside the classroom, supported by a conversational application.
  • To support the learning of Italian L2 (pre-A1 e A1) by a segment of the population often overlooked by current applications, specifically people with little (if any) alphabetical, linguistic and meta-linguistic skills.
  • To support the work of teachers in the classroom, both in providing explanations to students and in the customization of individual exercises of language use.


  • The incremental development of a chatbot calibrated to the needs of Italian L2 learners from the pre-Alpha level to A2 level.
  • The incremental development of a simplified glossary, adapted to the above levels and enriched by multimedia components such as images, animations, videos and audio that the chatbot can call on to facilitate the language learning process.
  • The pilot testing of the chatbot in all classes held at the two Genoese CPIAs (Provincial Centres for the Education of Adults) involving about one hundred migrants.
  • The active involvement, in the research-action process, of about ten CPIA teachers, who will be involved in the following: supporting the CNR-ITD researchers in defining the functional specifications of the chatbot; Identification/formulation of everyday situations and related conversational routines; classroom beta-testing of the system with students; incremental development of the different components of the chatbot and the multimedia glossary.

Expected results

At the didactic-methodological level

The expected results at the didactic level can be traced back to the research questions that the project intends to address, that is:

  1. Significant improvements in the learning process (when compared with the use of traditional tools) fostered by student interaction with a chatbot through conversational routines focused on exploring the terms contained in an integrated multimedia glossary and situations typical of everyday life. These expected improvements relate to the achievement of the competence objectives defined by the CEFR and the reference syllabuses for the levels identified.
  2. Lightening the teacher’s classroom teaching burden (in the presentation and exercise phases) via use of a hybrid learning environment. It should be remembered that the hybridization is at the level of both space (classroom, extra-aula, digital), and didactic mediation (teacher in synergy with the chatbot). It is assumed that the tool will constitute an agile resource:
    • for teachers to find simplified contents regarding the lexicon dealt with during the lesson (interactive multimedia glossary);
    • to support students' literacy strategies, which can be easily used by the students themselves (functions of speech transcription and vocal synthesis of written text);
    • that is automated, thus allowing, within the lesson, interaction customized to each student’s needs during the phases of autonomous exercise (exercises included in the glossary).
  3. Higher level of learner engagement during the lesson and stimulus to continue language practice outside the classroom, with activities that the students consider significant and relevant to their life experiences (specifically, interaction in natural language with an automatic agent that re-proposes real situations) and during which they are stimulated to act with the language in different contexts.

At the technological level

The expected technological results are as follows:

  • development of a fully functional, multimodal/multisurface chatbot, based on artificial intelligence (AI) techniques, capable of interacting vocally and in writing, with examples supported by multimedia content, and the capability to generate language learning tests in a playful way. The chatbot must be:
    • both reactive ("pull" mode), answering questions from the learner, and proactive ("push" mode), stimulating the learner with actions to be carried out, such as exercises and tests, partly on the basis of previous interactions;
    • developed incrementally, starting from the elementary functions for exploring the terms of the glossary and then moving on to mini-dialogues and finally to situated dialogues, i.e. immersed in a possible daily situation (buying, working, etc.);
    • easy for students to (learn how to) use and provide strong incentive for use inside and outside the classroom;
  • a simplified glossary accompanying the chatbot, adapted to the above levels, enriched with multimedia components (images, animations, video and audio) to facilitate/enrich interaction with the conversational application. The multimedia glossary accompanying the chatbot, to be co-constructed with the CPIA teachers based on their professional experience, will be made available in crowdsourcing mode to encourage incremental development by anyone willing to collaborate in its implementation and, consequently, its use.

Project e-mail: edubots@itd.cnr.it


  • Doctoral school in “Digital Humanities” at the University of Genoa
  • Department of Modern Languages and Culture (DLCM) at the University of Genoa
  • Genoa Provincial Centres for the Education of Adults