metadata
dataset_info:
features:
- name: audio_id
dtype: string
- name: word_index_in_sentence
dtype: int64
- name: word_index_in_word_ctm
dtype: int64
- name: buckwalter
dtype: string
- name: word_start
dtype: float64
- name: word_duration
dtype: float64
- name: word_end
dtype: float64
- name: word
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 177581012.72
num_examples: 14534
download_size: 176361209
dataset_size: 177581012.72
configs:
- config_name: default
data_files:
- split: train
path: data/train-*