metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: duration
dtype: float64
- name: offset
dtype: float64
- name: speaker_id
dtype: string
- name: sentence
dtype: string
splits:
- name: validation
num_bytes: 119191914
num_examples: 125
- name: test
num_bytes: 120009795
num_examples: 125
- name: train
num_bytes: 134161016.88481675
num_examples: 399
download_size: 387815841
dataset_size: 373362725.88481677
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*