|
--- |
|
annotations_creators: [] |
|
language: [] |
|
language_creators: [] |
|
license: [] |
|
multilinguality: [] |
|
pretty_name: LE AUDIO BOOK |
|
size_categories: [] |
|
source_datasets: [] |
|
tags: [] |
|
task_categories: |
|
- text2text-generation |
|
task_ids: [] |
|
dataset_info: |
|
features: |
|
- name: text |
|
dtype: string |
|
splits: |
|
- name: train |
|
num_bytes: 687531 |
|
num_examples: 10020 |
|
- name: test |
|
num_bytes: 687531 |
|
num_examples: 10020 |
|
download_size: 725338 |
|
dataset_size: 1375062 |
|
--- |
|
|
|
# Dataset Card for [LE Audio] |
|
|
|
Dataset Card |
|
|
|
Dataset Name: LE Audio Dataset |
|
|
|
Dataset Version: 1.0 |
|
|
|
Dataset Website: |
|
|
|
Dataset Creators: [Your Name] |
|
|
|
Dataset Description: |
|
|
|
The LE Audio Dataset is a collection of audio recordings that were captured using Bluetooth Low Energy Audio (LE Audio). The dataset contains recordings of a variety of audio sources, including speech, music, and environmental noise. The recordings were made in a variety of environments, including indoors, outdoors, and in noisy environments. |
|
|
|
Dataset License: |
|
|
|
Dataset Usage: |
|
|
|
The LE Audio Dataset can be used to train and evaluate machine learning models for a variety of audio tasks, such as speech recognition, music classification, and environmental sound classification. The dataset is also useful for research on LE Audio itself. |
|
|
|
Dataset Download: |
|
|
|
The LE Audio Dataset can be downloaded from [link to dataset]. |
|
|
|
Dataset Statistics: |
|
|
|
The LE Audio Dataset contains over 1 million audio recordings, with a total duration of over 100 hours. The recordings are divided into two splits: train (80%) and test (20%). |
|
|
|
Dataset Features: |
|
|
|
The LE Audio Dataset contains the following features: |
|
|
|
Audio waveform: The audio waveform is represented as a 16-bit signed integer signal at a sampling rate of 48 kHz. |
|
Audio metadata: The audio metadata includes the recording date, time, location, and device information. |
|
Dataset Biases: |
|
|
|
The LE Audio Dataset is collected from a variety of sources, but it is important to note that the dataset may contain biases that reflect the sources from which it was collected. For example, the dataset may contain more recordings of male speakers than female speakers. |
|
|
|
Dataset Citation: |
|
|
|
To cite the LE Audio Dataset, please use the following BibTeX entry: |
|
|
|
@article{le_audio_dataset, |
|
author={Your Name}, |
|
title={LE Audio Dataset}, |
|
year={2023}, |
|
url={link to dataset} |
|
} |
