This repository provides dataset of the following paper:
KEBAP: Korean Error Explainable Benchmark Dataset for ASR and Post-processing
- Seonmin Koo* , Chanjun Park*, Jinsung Kim, Jaehyung Seo, Sugyeong Eo, Hyeonseok Moon, Heuiseok Lim
Our KEBAP dataset can be found in data/
. Please refer to KEBAP paper for the detail of the dataset.
Please download the 'sound' folder from the provided link and place them in the directory /data/sound.
@inproceedings{
koo2023kebap,
title={{KEBAP}: Korean Error Explainable Benchmark Dataset for {ASR} and Post-processing},
author={Seonmin Koo, Chanjun Park, Jinsung Kim, Jaehyung Seo, Sugyeong Eo, Hyeonseok Moon, and Heuiseok Lim},
booktitle={The 2023 Conference on Empirical Methods in Natural Language Processing},
year={2023},
url={https://openreview.net/forum?id=i17SCD0YDI}
}