metadata
dataset_info:
config_name: it
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: 28848023139.798
num_examples: 33327
- name: validation
num_bytes: 2920073390.718
num_examples: 3422
- name: test
num_bytes: 3052236346.46
num_examples: 3595
download_size: 32347254190
dataset_size: 34820332876.976
configs:
- config_name: it
data_files:
- split: train
path: it/train-*
- split: validation
path: it/validation-*
- split: test
path: it/test-*