metadata
dataset_info:
features:
- name: audio_id
dtype: string
- name: speaker_id
dtype: string
- name: country
dtype: string
- name: gender
dtype: string
- name: duration
dtype: float64
- name: normalized_text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 7948285.2
num_examples: 4
download_size: 9017615
dataset_size: 7948285.2
configs:
- config_name: default
data_files:
- split: train
path: data/train-*