Under the guidance of Professor Changwon Lim from the Department of Statistics and Data Science, AISTAT Lab members Hyunjun Kim, Hyungyong Choi, Kyuwon Choi, and Eunsin Choi won first place in the DCASE 2025 Language-Based Audio Retrieval competition.

The DCASE 2025 Language-Based Audio Retrieval task is a cross-modal retrieval AI challenge where participants use human-written audio captions as text queries to retrieve the corresponding audio clips from the large-scale Clotho v2 dataset, with performance evaluated using accuracy metrics (e.g., mAP@16).

For more details, please visit: https://dcase.community/challenge2025/task-language-based-audio-retrieval-results.