With the advancement of technology, internet usage has surged and brought most global languages into the digital age. However, many regional and under-resourced languages continue to face challenges due to the lack of progress in language technology. One such group of languages is the Dravidian language family primarily spoken in South India and Sri Lanka, with communities of speakers also found in Nepal, Pakistan,Malaysia, London, other parts of India, and various regions around the world. Dravidian languages, which are 4,500 years old and spoken by millions across the world, remain underrepresented in digital spaces. This is the fifth workshop on speech and language technologies for Dravidian languages.
The main objectives of DravidianLangTech-2025 are as follows:
To identify and address the challenges involved in developing speech and language resources for dravidian languages.
To implement suitable language technology models tailored to the unique needs of dravidian languages.
To promote cross-disciplinary collaboration among global researchers.
To promote inclusive and safe digital spaces for dravidian language speakers through multimodal content moderation.
Call for Papers :
DravidianLangTech-2025 welcomes theoretical and practical paper submission on any Dravidian languages (Tamil, Kannada, Malayalam, Telugu, Tulu, Allar, Aranadan, Attapadya, Kurumba, Badaga, Beary, Betta Kurumba, Bharia, Bishavan, Brahui, Chenchu, Duruwa, Eravallan, Gondi, Holiya, Irula, Jeseri, Kadar, Kaikadi, Kalanadi, Kanikkaran, Khiwar, Kodava, Kolami, Konda, Koraga, Kota, Koya, Kurambhag Paharia, Kui, Kumbaran, Kunduvadi, Kurichiya, Kurukh, Kurumba, Kuvi, Madiya, Mala Malasar, Malankuravan, Malapandaram, Malasar, Malto, Manda, Muduga, Mullu Kurumba, Muria, Muthuvan, Naiki, Ollari, Paliyan, Paniya, Pardhan, Pathiya, Pattapu, Pengo, Ravula, Sholaga, Thachanadan, Toda, Wayanad Chetti, and Yerukala) that contributes to research in language processing, speech technologies or resources for the same. We will particularly encourage studies that address either practical application or improving resources for a given language in the field.
This comprehensive call for papers invites submissions on critical topics related to hate speech, offensive language, misinformation, and content safety in social media and online environments, especially as they pertain to Dravidian languages. Researchers are encouraged to explore innovative methods for detecting and mitigating various forms of harmful content, such as political hate speech, offensive language, and AI-generated misinformation, with a focus on cross-lingual and multimodal approaches that integrate text, images, and video.
Topics of interest include, but are not limited to:
Detection of Political Hate Speech in social media
Multimodal Hate Speech Detection Across Text, Image, and Video Content
AI-Generated Content Detection and Mitigation in Online Media
Corpus Development for Hate Speech Detection in Multilingual Contexts
Identifying Offensive Language in Political Discourse on Social Platforms
Cross-Modal Techniques for Multimodal Hate Speech Detection
Fake News and Rumor Detection in Dravidian Language Media
Emotion Analysis and Sentiment Detection in Hate Speech
Cyberbullying and Hostility Detection for Safer Online Environments
Disinformation and Misinformation Detection in Political Speech
Racial and Religious Abuse Detection Using Multimodal Approaches
Automated Detection of AI-Generated Harmful Content
Social Contagion of Hate Speech in Online Political Discussions
Accent and Emotion Recognition in Dravidian Language Speech
Social Bias Detection in AI-Generated Text
Sexism and Misogynistic Attitudes in Multimodal Online Content
Detection of Violent Incidents in social media Using Multimodal Data
Phonology and Morphology in Dravidian Language Processing
Document and Image Analysis for Hate Speech Detection in social media