metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: sentence
dtype: string
- name: accents
dtype: string
- name: Transcript
dtype: string
- name: CanonicalOld
dtype: string
- name: Canonical
dtype: string
- name: Error
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 3576524484.25
num_examples: 17582
download_size: 3215426222
dataset_size: 3576524484.25
configs:
- config_name: default
data_files:
- split: train
path: data/train-*