An Ear to the Sky: Intuitive Exploration & Discovery in Astronomical Data using Sonification

倾听天空：直觉探索

基本信息

批准号：
ST/X004651/1
负责人：
James Trayford
金额：
$ 55.22万
依托单位：
University of Portsmouth
依托单位国家：
英国
项目类别：
Research Grant
财政年份：
2023
资助国家：
英国
起止时间：
2023 至无数据
项目状态：
未结题

来源：
https://gtr.ukri.org/projects?ref=ST%2FX004651%2F1
关键词：
Ear Sky Intuitive Exploration Discovery

项目摘要

The recent decades of astronomy has seen an ever more detailed physical picture of our Universe emerge, with the enhanced sophistication of observations and physical models. However, the growing detail and volume of data presents its own problems; human data exploration can become overwhelming, despite its central role in historical scientific discovery. Machine-learning tools have arisen to help, but human intuition for the data remains key to their development and interpretation. This problem afflicts how we interface with data at all levels, from high-level research, to presenting data to the public. What's more, the visual bias of data presentation excludes those with visual impairments, likely contributing to a strong underrepresentation of Visually Impaired (VI) researchers in the field. For this program we propose development of novel tools and techniques for 'sonification' (encoding data into sound) to provide a unique perspective on astronomical data sets, in particular the STRAUSS code (Sonification Tools and Resources for Analysis Using Sound Synthesis). This free and open source python library uses an object-oriented and abstracted approach to allow uses to 'sonify' their own data, analogous to how a plotting library can be used for visualisation. STRAUSS will be used to sonify light spectra for classification, and as the basis for a novel IFU explorer, allowing users to hear the third frequency axis as they explore the 2D-projected spatial axes. We will also explore using STRAUSS to monitor high-volume data and as a way to present data to the public via a unique planetarium & VR show. Together, we intend to partner with commercial firms interested in human curation and exploration of big-data, as well as accessibility for a more inclusive society. Through commercial support we hope to maximise the broad societal and interdisciplinary benefits we perceive, establishing new intuition and accessibility for complex data in the global community.

近几十年来，随着观测和物理模型的日益复杂，天文学已经出现了更加详细的宇宙物理图景。然而，不断增长的数据细节和数量也带来了其自身的问题；尽管人类数据探索在历史科学发现中发挥着核心作用，但它可能会变得势不可挡。机器学习工具的出现是为了提供帮助，但人类对数据的直觉仍然是其开发和解释的关键。这个问题影响着我们与各个层面的数据交互的方式，从高级研究到向公众展示数据。更重要的是，数据呈现的视觉偏差将那些有视觉障碍的人排除在外，这可能导致该领域的视觉障碍（VI）研究人员代表性严重不足。对于这个项目，我们建议开发新的“声音化”工具和技术（将数据编码为声音），以提供对天文数据集的独特视角，特别是STRAUSS代码（使用声音合成进行分析的声音化工具和资源）。这个免费开源的 python 库使用面向对象和抽象的方法，允许用户“声音化”自己的数据，类似于绘图库用于可视化的方式。 STAUSS 将用于对光谱进行声化以进行分类，并作为新型 IFU 浏览器的基础，使用户在探索 2D 投影空间轴时能够听到第三频率轴。我们还将探索使用 STRAUSS 来监测大量数据，并通过独特的天文馆和 VR 表演向公众展示数据。我们打算与对人类管理和大数据探索以及更包容的社会的可访问性感兴趣的商业公司合作。通过商业支持，我们希望最大限度地发挥我们所感知到的广泛社会和跨学科效益，为全球社区的复杂数据建立新的直觉和可访问性。