metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: transcription
dtype: string
- name: transcription_normalized
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 1769663213.731
num_examples: 3123
download_size: 1591894082
dataset_size: 1769663213.731
configs:
- config_name: default
data_files:
- split: train
path: data/train-*