metadata
dataset_info:
features:
- name: audio
struct:
- name: array
sequence: float32
- name: sampling_rate
dtype: int64
- name: transcription
dtype: string
splits:
- name: train
num_bytes: 2043930336
num_examples: 7200
download_size: 2050079624
dataset_size: 2043930336
configs:
- config_name: default
data_files:
- split: train
path: data/train-*