metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
- name: speaker
dtype: string
- name: chunk_id
dtype: string
splits:
- name: train
num_bytes: 23935066611
num_examples: 2881
download_size: 22278400242
dataset_size: 23935066611
configs:
- config_name: default
data_files:
- split: train
path: data/train-*