metadata
dataset_info:
features:
- name: path
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 48000
- name: transcription
dtype: string
- name: normalized_text
dtype: string
splits:
- name: train
num_bytes: 222368534
num_examples: 5000
download_size: 218870883
dataset_size: 222368534
configs:
- config_name: default
data_files:
- split: train
path: data/train-*