metadata
dataset_info:
features:
- name: audio
struct:
- name: array
sequence: float32
- name: sampling_rate
dtype: int64
- name: sentence
dtype: string
- name: path
dtype: string
- name: gender
dtype: string
- name: accent
dtype: string
- name: speaker
dtype: string
splits:
- name: train
num_bytes: 7154210688
num_examples: 31394
download_size: 6724197122
dataset_size: 7154210688
configs:
- config_name: default
data_files:
- split: train
path: data/train-*