metadata
dataset_info:
features:
- name: sentence_norm
dtype: string
- name: audio
struct:
- name: array
sequence: int64
- name: path
dtype: string
- name: sampling_rate
dtype: int64
- name: wer
dtype: int64
- name: id
dtype: string
splits:
- name: train
num_bytes: 212990169
num_examples: 498
download_size: 47949623
dataset_size: 212990169
configs:
- config_name: default
data_files:
- split: train
path: data/train-*