metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: filename
dtype: string
- name: word
dtype: string
- name: phoneme
dtype: string
- name: phonemeCount
dtype: int64
splits:
- name: test
num_bytes: 663350436.601
num_examples: 45617
download_size: 310648963
dataset_size: 663350436.601
configs:
- config_name: default
data_files:
- split: test
path: data/test-*