metadata
dataset_info:
features:
- name: audio
struct:
- name: array
sequence: float32
- name: sampling_rate
dtype: int64
splits:
- name: train
num_bytes: 1537885380
num_examples: 12950
download_size: 773608841
dataset_size: 1537885380
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
language:
- ru