metadata
dataset_info:
features:
- name: speech
sequence: float64
- name: phoneme
dtype: string
- name: transcription
dtype: string
- name: speaker
dtype: string
- name: speaker ID
dtype: int64
- name: language
dtype: string
splits:
- name: test
num_bytes: 42948245
num_examples: 20
download_size: 37246766
dataset_size: 42948245
configs:
- config_name: default
data_files:
- split: test
path: data/test-*