metadata
dataset_info:
features:
- name: audio
dtype:
audio:
decode: false
- name: text
dtype: string
- name: transcription_corrigee
dtype: string
- name: orthography
dtype: string
- name: source
dtype: string
- name: duration
dtype: string
- name: source_record_id
dtype: string
splits:
- name: train
num_bytes: 383423
num_examples: 5
download_size: 380030
dataset_size: 383423
configs:
- config_name: default
data_files:
- split: train
path: data/train-*