Datasets:
metadata
dataset_info:
features:
- name: duration
dtype: float64
- name: audio_filepath
dtype: string
- name: text
dtype: string
- name: audio
dtype: audio
splits:
- name: train
num_bytes: 14999041787.472
num_examples: 79179
- name: test
num_bytes: 6596357844.232
num_examples: 32946
download_size: 17857403377
dataset_size: 21595399631.704002
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*
task_categories:
- translation
language:
- es
- en
tags:
- speech-translation