metadata
dataset_info:
features:
- name: audio_path
dtype: string
- name: buckwalter_text
dtype: string
- name: buckwalter_text_numbers
dtype: string
- name: arabic_text
dtype: string
- name: arabic_text_numbers
dtype: string
- name: duration
dtype: float64
splits:
- name: train
num_bytes: 8564533
num_examples: 13588
- name: validation
num_bytes: 4653534
num_examples: 7999
- name: test
num_bytes: 13097054
num_examples: 20818
download_size: 8765593
dataset_size: 26315121
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*