metadata
dataset_info:
config_name: eu
features:
- name: path
dtype: string
- name: audio
dtype:
audio:
sampling_rate: 16000
- name: sentence
dtype: string
- name: condition_on_prev
sequence: int64
- name: whisper_transcript
dtype: string
splits:
- name: train
num_bytes: 13429091230.826
num_examples: 15549
- name: validation
num_bytes: 2595918690.444
num_examples: 3028
- name: test
num_bytes: 2654049843.935
num_examples: 3095
download_size: 16365308515
dataset_size: 18679059765.205
configs:
- config_name: eu
data_files:
- split: train
path: eu/train-*
- split: validation
path: eu/validation-*
- split: test
path: eu/test-*