|
|
--- |
|
|
dataset_info: |
|
|
features: |
|
|
- name: audio |
|
|
dtype: |
|
|
audio: |
|
|
sampling_rate: 16000 |
|
|
- name: text |
|
|
dtype: string |
|
|
splits: |
|
|
- name: train |
|
|
num_bytes: 1117294900.769 |
|
|
num_examples: 10227 |
|
|
download_size: 1021497255 |
|
|
dataset_size: 1117294900.769 |
|
|
configs: |
|
|
- config_name: default |
|
|
data_files: |
|
|
- split: train |
|
|
path: data/train-* |
|
|
--- |
|
|
# Zeroth-STT-Korean Dataset |
|
|
|
|
|
## Description |
|
|
This is a one of ten equally divided parts of the Zeroth-STT-Korean dataset. Ten equal shards of the original dataset were created and uploaded as individual repos for fine-tuning/training needs. |
|
|
|
|
|
## Citation |
|
|
|
|
|
Zeroth-Korean Dataset, created by [Lucas Jo(@Atlas Guide Inc.) and Wonkyum Lee(@Gridspace Inc.)], 2023. |
|
|
Available at https://github.com/goodatlas/zeroth under CC-BY-4.0 license. |
|
|
|
|
|
Junhoee/STT_Korean_Dataset_80000 Dataset, created by [Junhoee], 2024. |
|
|
Available at https://huggingface.co/datasets/Junhoee/STT_Korean_Dataset_80000 |
