metadata
dataset_info:
features:
- name: file
dtype: string
- name: audio
dtype: audio
- name: utterance
dtype: string
- name: speaker
dtype: string
- name: context
sequence: string
- name: context_speakers
sequence: string
- name: show
dtype: string
- name: label
dtype: bool
- name: instruction
dtype: string
- name: instruction_TTS
dtype:
audio:
sampling_rate: 16000
splits:
- name: test
num_bytes: 80755183
num_examples: 200
download_size: 36370726
dataset_size: 80755183
configs:
- config_name: default
data_files:
- split: test
path: data/test-*