Datasets:
metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: language
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: prediction
dtype: string
splits:
- name: test
num_bytes: 954235048.75
num_examples: 5386
download_size: 821835926
dataset_size: 954235048.75
configs:
- config_name: default
data_files:
- split: test
path: data/test-*