metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: transcription
dtype: string
- name: transcription_normalised
dtype: string
- name: phonemes
dtype: string
- name: gender
dtype: string
splits:
- name: train
num_bytes: 6917771782
num_examples: 7525
download_size: 6796578625
dataset_size: 6917771782
configs:
- config_name: default
data_files:
- split: train
path: data/train-*