metadata
dataset_info:
config_name: fr
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: 92381810563.144
num_examples: 106356
- name: validation
num_bytes: 3020318043
num_examples: 3520
- name: test
num_bytes: 3013639822.706
num_examples: 3522
download_size: 90949907674
dataset_size: 98415768428.84999
configs:
- config_name: fr
data_files:
- split: train
path: fr/train-*
- split: validation
path: fr/validation-*
- split: test
path: fr/test-*