metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: text
dtype: string
- name: duration
dtype: float64
- name: ratio
dtype: float64
- name: videoid
dtype: string
- name: key
dtype: string
- name: dataset_id
dtype: string
- name: lang
dtype: 'null'
- name: text_en_gpt3.5
dtype: string
- name: audio_en_openai
dtype: audio
- name: audio_tokens_speed_up
sequence:
sequence: int64
- name: audio_tokens
sequence:
sequence: int64
splits:
- name: train
num_bytes: 24569011
num_examples: 100
download_size: 22122484
dataset_size: 24569011
configs:
- config_name: default
data_files:
- split: train
path: data/train-*