metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: text
dtype: string
- name: normalized_text
dtype: string
- name: speaker
dtype: string
splits:
- name: test
num_bytes: 7993646
num_examples: 95
- name: train
num_bytes: 68421127
num_examples: 277
download_size: 76394317
dataset_size: 76414773
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
- split: train
path: data/train-*