CLAP features of CASTELLA

What is this? This repository contains audio and text features of CASTELLA dataset extracted by MS-CLAP 2023. Using these features, we can reproduce the audio moments retrieval using CASTELLA, which is used in lighthouse. Please also check demo page. File structure features/ └──castella ├──clap │ └──{yid}.npz └──clap_text └──qid_{yid}_{caption_id}.npz How to Use on Lighthouse? The `.tar.gz` files should be decompressed by following shell commands:tar -zxvf clap.tar.gz -C {LIGHTHOUSE_PATH}/features/castellatar -zxvf clap_text.tar.gz -C {LIGHTHOUSE_PATH}/features/castella Version ver.2: revise an error of G_Xod6VFxNY_2 Citation @article{munakata2025castella, title={CASTELLA: Long Audio Dataset with Captions and Temporal Boundaries}, author={Munakata, Hokuto and Takehiro, Imamura and Nishimura, Taichi and Komatsu, Tatsuya}, journal={arXiv preprint arXiv:2511.15131}, year={2025},}