metadata
dataset_info:
features:
- name: source
dtype: string
- name: caption
dtype: string
- name: text
dtype: string
- name: speech_duration
dtype: float64
- name: accent
dtype: string
- name: emotion
dtype: string
- name: speaking_rate
dtype: string
- name: intrinsic_tags
list: string
- name: emotion_text_prompt
dtype: string
- name: all_tags
list: string
- name: age
dtype: string
- name: pitch
dtype: string
- name: situational_tags
list: string
- name: gender
dtype: string
- name: basic_tags
list: string
- name: noise
dtype: string
- name: speech_monotony
dtype: string
- name: clean_caption
dtype: string
- name: rebuilt_caption
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 44100
splits:
- name: train
num_bytes: 41379818009.746
num_examples: 14159
download_size: 40701400619
dataset_size: 41379818009.746
configs:
- config_name: default
data_files:
- split: train
path: data/train-*