metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: sentence
dtype: string
- name: age
dtype: string
- name: gender
dtype: string
- name: accents
dtype: string
- name: Transcript
dtype: string
- name: Canonical
dtype: string
- name: Canonical_Sep
dtype: string
- name: Error
dtype: string
- name: CanonicalOld
dtype: string
- name: duration
dtype: float64
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 725440737.0293732
num_examples: 18585
- name: test
num_bytes: 3978250.992813415
num_examples: 100
- name: val
num_bytes: 3843789.992813415
num_examples: 100
download_size: 702688174
dataset_size: 733262778.0150001
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
- split: val
path: data/val-*