metadata
license: mit
dataset_info:
features:
- name: audio
dtype: audio
- name: transcript
dtype: string
- name: normalized_transcript
dtype: string
- name: speaker_id
dtype: int64
splits:
- name: train
num_bytes: 182446597
num_examples: 293
- name: val
num_bytes: 24689960
num_examples: 33
download_size: 204585931
dataset_size: 207136557
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: val
path: data/val-*