metadata
dataset_info:
features:
- name: audio
dtype: audio
- name: transcription
dtype: string
- name: transcription_language
dtype: string
- name: detected_spoken_language
dtype: string
- name: audio_id
dtype: string
splits:
- name: train
num_bytes: 98743
num_examples: 3
download_size: 102068
dataset_size: 98743
configs:
- config_name: default
data_files:
- split: train
path: data/train-*