metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: file_path
dtype: string
- name: topic
dtype: string
- name: darija
dtype: string
- name: english
dtype: string
- name: arabizi
dtype: string
- name: subtopic
dtype: string
splits:
- name: train
num_bytes: 40391504
num_examples: 694
download_size: 40329967
dataset_size: 40391504
configs:
- config_name: default
data_files:
- split: train
path: data/train-*