metadata
dataset_info:
features:
- name: id
dtype: string
- name: type
dtype: string
- name: full_audio
struct:
- name: path
dtype: string
- name: bytes
dtype: binary
- name: speaking_segments
list:
- name: path
dtype: string
- name: bytes
dtype: binary
- name: speaking_transcriptions
list:
- name: path
dtype: string
- name: text
dtype: string
- name: non_speaking_segments
list:
- name: path
dtype: string
- name: bytes
dtype: binary
- name: non_speaking_transcriptions
list:
- name: path
dtype: string
- name: text
dtype: string
splits:
- name: train
num_bytes: 18680358823
num_examples: 924
download_size: 18681674300
dataset_size: 18680358823
configs:
- config_name: default
data_files:
- split: train
path: data/train-*