metadata
dataset_info:
features:
- name: path
dtype: string
- name: sentence
dtype: string
- name: utt_id
dtype: string
- name: segment
struct:
- name: duration
dtype: float64
- name: end
dtype: float64
- name: rec_id
dtype: string
- name: start
dtype: float64
- name: audio
sequence: float32
splits:
- name: train
num_bytes: 2829847729
num_examples: 10841
download_size: 1296978579
dataset_size: 2829847729
configs:
- config_name: default
data_files:
- split: train
path: data/train-*