metadata
dataset_info:
features:
- name: ID
dtype: string
- name: audio
dtype: audio
- name: text
dtype: string
- name: text_phowhisper
dtype: string
- name: speaker_accent
dtype: string
- name: speaker_gender
dtype: string
- name: speaker_name
dtype: string
- name: duration
dtype: float64
- name: video_id
dtype: string
- name: audio_file
dtype: string
splits:
- name: train
num_bytes: 71220788885.434
num_examples: 236078
download_size: 60680774232
dataset_size: 71220788885.434
configs:
- config_name: default
data_files:
- split: train
path: data/train-*