metadata
dataset_info:
features:
- name: id
dtype: string
- name: normalized_text
dtype: string
- name: speaker_id
dtype: int32
- name: split
dtype: string
- name: sentiment
dtype: string
- name: start_second
dtype: float32
- name: end_second
dtype: float32
- name: audio_path
dtype: string
config_name: voxceleb
splits:
- name: train
num_bytes: 2287456
num_examples: 5777
- name: validation
num_bytes: 551279
num_examples: 1454
- name: test
num_bytes: 889140
num_examples: 3553
download_size: 1545458156
dataset_size: 3727875