metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: norm_text
dtype: string
- name: file_name
dtype: string
- name: original_media
dtype: string
- name: segment_id
dtype: int64
- name: avg_score
dtype: float64
- name: num_words
dtype: int64
- name: duration
dtype: float64
- name: is_valid_duration
dtype: bool
- name: cps
dtype: float64
- name: num_chars
dtype: int64
splits:
- name: train
num_bytes: 10319445267
num_examples: 9403
download_size: 7257538008
dataset_size: 10319445267
configs:
- config_name: default
data_files:
- split: train
path: data/train-*