|
|
--- |
|
|
dataset_info: |
|
|
features: |
|
|
- name: uid |
|
|
dtype: string |
|
|
- name: path |
|
|
dtype: string |
|
|
- name: start |
|
|
dtype: string |
|
|
- name: end |
|
|
dtype: string |
|
|
- name: text |
|
|
dtype: string |
|
|
- name: audio |
|
|
dtype: audio |
|
|
splits: |
|
|
- name: train |
|
|
num_bytes: 12098819649.84326 |
|
|
num_examples: 326620 |
|
|
download_size: 11979679763 |
|
|
dataset_size: 12098819649.84326 |
|
|
configs: |
|
|
- config_name: default |
|
|
data_files: |
|
|
- split: train |
|
|
path: data/train-* |
|
|
--- |
|
|
|
|
|
This is a large Persian ASR dataset gathered from carefully extracting the subtitles from a couple of series and shows on the Filimo platform by the Playto team at SabaIdea. |
|
|
|
|
|
We will probably make this dataset public in the near future. |
|
|
|
