metadata
dataset_info:
features:
- name: AUDIO_ID
dtype: string
- name: SCORE
dtype: string
- name: FORM
dtype: string
- name: ASR
dtype: string
- name: TEXT
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train.75
num_bytes: 167224573.76190478
num_examples: 62
- name: test.25
num_bytes: 58637292.23809524
num_examples: 22
download_size: 224014676
dataset_size: 225861866
configs:
- config_name: default
data_files:
- split: train.75
path: data/train.75-*
- split: test.25
path: data/test.25-*