Multilingual approaches to dialogue systems
and Conversational AI (MACAI-21)

A special session at IWSDS-2021

Call for papers

Multilinguality is one of the key challenges in spoken human-machine interaction. Popular voice assistants such as Alexa and Google Assistant offer interaction in a limited number of languages. However, their approaches often rely on language-specific technology development. Research endeavors are even more limited as they mostly focus solely on English, e.g., the popular model GPT-3 is not available in any other language. Furthermore, most available dialogue research corpora only focus on English or Chinese. However, only less than 5 % of the world population have English as their first language (less than 12 % have Chinese) and among the top 10 languages, many are not represented by today's dialogue systems.

Voice, though, is one of the primary means of communication and provides access not only to commercial services but also to humanitarian and other non-profit services. These dialogue systems are required to converse in many different languages which needs to be reflected in research and development of such systems.

The goal of this workshop is to bring together researchers from the different disciplines (incl. NLP, HCI, Dialogue) to give them a forum to exchange ideas and identify common problems, to create synergy effects, and generally to advance the field of multilingual processing. Invited are all submissions that target structured approaches related to multilingual dialogue systems and their individual components. This includes, but is not limited to, the following topics:

Submission information

Submission are managed together with IWSDS submissions. More information on submissions is available at the IWSDS website.

Important dates