metadata
dataset_info:
features:
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: major_emotion
dtype: string
- name: transcription
dtype: string
- name: prompt
dtype: string
- name: category
dtype: string
- name: label
dtype: string
- name: transcript
dtype: string
- name: input_ids
sequence: int32
splits:
- name: test
num_bytes: 304817119.53
num_examples: 2170
download_size: 299258232
dataset_size: 304817119.53
configs:
- config_name: default
data_files:
- split: test
path: data/test-*