metadata
dataset_info:
features:
- name: id
dtype: string
- name: file
dtype: string
- name: text
dtype: string
- name: normalized_text
dtype: string
- name: audio_token
sequence:
sequence:
sequence: int64
splits:
- name: train
num_bytes: 57110018
num_examples: 13100
download_size: 11862590
dataset_size: 57110018
configs:
- config_name: default
data_files:
- split: train
path: data/train-*