metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: target
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
splits:
- name: train
num_bytes: 46553190
num_examples: 520
download_size: 46412864
dataset_size: 46553190
configs:
- config_name: default
data_files:
- split: train
path: data/train-*