metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: gender
dtype: string
- name: accent
dtype: string
- name: speaker_id
dtype: string
- name: text
dtype: string
- name: speaker_name
dtype: string
- name: normalized_text
dtype: string
splits:
- name: train
num_bytes: 9474956225.220339
num_examples: 69486
- name: validation
num_bytes: 1184393333.3718305
num_examples: 8686
- name: test
num_bytes: 1184398692.6338303
num_examples: 8686
download_size: 11234963565
dataset_size: 11843748251.226
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*
- split: test
path: data/test-*