metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
- name: raw_transcription
dtype: string
- name: speaker_id
dtype: int64
splits:
- name: train
num_bytes: 28833809.002
num_examples: 2058
download_size: 27249237
dataset_size: 28833809.002
configs:
- config_name: default
data_files:
- split: train
path: data/train-*