metadata
dataset_info:
features:
- name: speaker_id
dtype: string
- name: transcription_id
dtype: int64
- name: text
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 44100
splits:
- name: train
num_bytes: 5526903541.580767
num_examples: 9521
- name: test
num_bytes: 11609922.364417113
num_examples: 20
download_size: 5043779526
dataset_size: 5538513463.945184
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*