metadata
dataset_info:
features:
- name: transcription
dtype: string
- name: audio
dtype: audio
- name: speaker
dtype: string
splits:
- name: train
num_bytes: 1465773956.24
num_examples: 2652
download_size: 1254041413
dataset_size: 1465773956.24
configs:
- config_name: default
data_files:
- split: train
path: data/train-*