metadata
dataset_info:
features:
- name: audio_path
dtype:
audio:
sampling_rate: 24000
- name: transcription
dtype: string
- name: language
dtype: string
- name: speaker_id
dtype: string
- name: gender
dtype: string
splits:
- name: train
num_bytes: 107419238340.4
num_examples: 462850
download_size: 118558735977
dataset_size: 107419238340.4
configs:
- config_name: default
data_files:
- split: train
path: data/train-*