metadata
dataset_info:
config_name: id
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: 886887591.415
num_examples: 1035
- name: validation
num_bytes: 445769434
num_examples: 505
- name: test
num_bytes: 479298192
num_examples: 542
download_size: 1688264749
dataset_size: 1811955217.415
configs:
- config_name: id
data_files:
- split: train
path: id/train-*
- split: validation
path: id/validation-*
- split: test
path: id/test-*