An Analysis of AI-Powered Sonification for the Blind and Visually Impaired    

An Analysis of AI-Powered Sonification for the Blind and Visually Impaired | Authorea try { document.documentElement.classList.add('js'); } catch (e) { } var _gaq = _gaq || []; _gaq.push(['_setAccount', 'G-8VDV14Y67G']); _gaq.push(['_trackPageview']); (function() { var ga = document.createElement('script'); ga.type = 'text/javascript'; ga.async = true; ga.src = ('https:' == document.location.protocol ? 'https://ssl' : 'http://www') + '.google-analytics.com/ga.js'; var s = document.getElementsByTagName('script')[0]; s.parentNode.insertBefore(ga, s); })(); Skip to main content Preprints Collections Wiley Open Research IET Open Research Ecological Society of Japan All Collections About About Authorea FAQs Contact Us Quick Search anywhere Search for preprint articles, keywords, etc. Search Search ADVANCED SEARCH SCROLL This is a preprint and has not been peer reviewed. Data may be preliminary. 4 August 2025 V1 Latest version Share on An Analysis of AI-Powered Sonification for the Blind and Visually Impaired Author : Anosha Dlair 0009-0008-9524-1187 [email protected] Authors Info & Affiliations https://doi.org/10.22541/au.175433669.98742423/v1 253 views 67 downloads Contents Abstract Information & Authors Metrics & Citations View Options References Figures Tables Media Share Abstract Astronomy is definitely beautiful, but what do you do with it when you can't see it?I've been thinking about this a lot lately what if your world isn't all about pictures, and what does it feel like to be a supernova? That question kept coming back to me. Along with the science, this article looks at how sound can help us see space. I look into the idea of using artificial intelligence to change sonification, which is the process of turning astronomical data into sound.I also look into how this could help people who are blind or have low vision. There are already some tools, like NASA's Astronify and xSonify . They are real, but they seem like the first steps. The actual possibility? It's in allowing AI to listen, learn, and adapt in addition to automating tasks. To create soundscapes from star maps and brightness curves that are both accurate and unique. And to be honest, I'm excited about that because it should never feel exclusive to be in space for anyone . It's not a tech review. It's a gentle prod, a viewpoint perhaps even a challenge. What if we began to hear the universe instead of attempting to " see"it? AI generated Data Input : Star attributes like brightness, coordinates, and pulsation patterns. patterns. Preprocessing: Normalize data, extract features. AI Model : Use a Transformer-based model (Hugging Face) to classify important features and personalize based on auditory profile. Sonification Module : Librosa used to convert key features into audio. For example, brightness = pitch, coordinate = stereo pan, pulsation = rhythm. Output: Rendered as a spatialized audio file playable through headphones or speakers. Pseudocode Snippet import librosa from transformers import pipeline # Load star data (brightness, RA/Dec, pulsation) star_data = load_star_dataset() # AI interprets significance model = pipeline(’text-classification’, model=’distilbert-base-uncased’) salient_features = model(star_data[’descriptions’]) # Map features to sound pitch = librosa.tone(frequency=brightness_to_pitch(star_data[’brightness’])) location = coordinate_to_stereo(star_data[’RA’], star_data[’Dec’]) rhythm = pulsation_to_rhythm(star_data[’pulsation’]) # Combine and export audio sonified_output = mix_audio(pitch, location, rhythm) export(sonified_output, filename=”star_sound.wav” Results / Proposed Evaluation A pilot test can be conducted with five visually impaired participants using three versions of the same star dataset: 1. AI-generated personalized sonification. 2. Standard NASA Astronify audio. 3. No audio, just verbal description. Participants will rate each on: - Emotional impact - Information clarity - Preference A Likert-scale survey and follow-up interviews will help identify strengths and limitations. Figure: AI-to-Sonification Pipeline Diagram [Star Dataset] –> [AI Feature Prioritization] –> [Personalized Mapping Rules] –> [Sound Engine (Librosa)] –> [Audio Output] Discussion Alright, so why haven’t we completed this task yet ? Why isn’t this more commonplace given how AI is improving daily, how much open data is available, and how many people are calling for more inclusive science? To be honest, some of it is just habit. The field of astronomy was born with images. We create charts, plot graphs, and observe galaxies. Some people find it strange to change that, as if you’re interfering with tradition. However if customs exclude people, they are not sacred. Another problem? Everybody is operating in their own bubble. Models are created by AI developers .Data is studied by astronomers. Accessibility professionals create fantastic resources.However, they don’t always come into contact. It’s unfortunate because when they work together, amazing things can occur. Funding comes next. Initiatives pertaining to accessibility or inclusion? They are frequently pushed aside. Not showy enough. Not ” core science. ” But shouldn’t our primary goal be to make science accessible to all? Additionally, and I’ll be honest, it’s difficult. It takes time to figure out how to translate complicated space data into understandable sound. It requires time. Patience is required. It takes individuals who are willing to make mistakes and grow from them. Nevertheless, I firmly feel that we are at a watershed. Using tools is now simpler than ever. People are begi nning to ask more insightful questions. Additionally, students like myself and perhaps you as well, are envisioning space as something you enter rather than something you observe from a distance. I wrote this because I kept thinking about what it might be like to sit in a planetarium or classroom and actually hear the beat of a galaxy. As experience as well as sound. to be amazed without having to see. To transform quiet into c ommunication. Perhaps that’s where it all starts. Not in an expensive lab. With imperfect code, no. One question, wh at if there is someone out there who is eager to hear the stars? Citations -Diaz-Merced, W. (2013). Sound for the visually impaired: A sonification tool for astronomy data. Harvard University. - Harvard University Astronomy Department. (n.d.). Astroacoustics Lab. - Hugging Face, n.d. Hugging Face: The future is being built by the AI community. taken from huggingface.co. - Librosa. (n.d.). Librosa is a library for analyzing audio in Python. - NASA. (n.d.). Astronomy for Kids from NASA. - https://science.nasa.gov/learners/astronify/ - NASA Goddard Space Flight Center. (n.d.). xSonify is a tool for turning data into sound. Found at https://science.gsfc.nasa.gov/sed/index.cfm?fuseAction=tools.main\&tool_id=22 “This paper was developed with assistance from AI for revising content.” Information & Authors Information Version history V1 Version 1 04 August 2025 Copyright This work is licensed under a Non Exclusive No Reuse License. Keywords artificial inteligence-ai assitive technology auditory display sonification visual impairment Authors Affiliations Anosha Dlair 0009-0008-9524-1187 [email protected] View all articles by this author Metrics & Citations Metrics Article Usage 253 views 67 downloads .FvxKWukQNSOunydq8rnd { width: 100px; } Citations Download citation Anosha Dlair. An Analysis of AI-Powered Sonification for the Blind and Visually Impaired . Authorea . 04 August 2025. DOI: https://doi.org/10.22541/au.175433669.98742423/v1 If you have the appropriate software installed, you can download article citation data to the citation manager of your choice. Simply select your manager software from the list below and click Download. For more information or tips please see 'Downloading to a citation manager' in the Help menu . Format Please select one from the list RIS (ProCite, Reference Manager) EndNote BibTex Medlars RefWorks Direct import Tips for downloading citations document.getElementById('citMgrHelpLink').addEventListener('click', function() { popupHelp(this.href); return false; }); $(".js__slcInclude").on("change", function(e){ if ($(this).val() == 'refworks') $('#direct').prop("checked", false); $('#direct').prop("disabled", ($(this).val() == 'refworks')); }); View Options View options PDF View PDF Figures Tables Media Share Share Share article link Copy Link Copied! Copying failed. Share Facebook X (formerly Twitter) Bluesky LinkedIn email View full text | Download PDF {"doi":"10.22541/au.175433669.98742423/v1","type":"Article"} Now Reading: Share Figures Tables Close figure viewer Back to article Figure title goes here Change zoom level Go to figure location within the article Download figure Toggle share panel Toggle share panel Share Toggle information panel Toggle information panel Go to previous graphic Go to next graphic Go to previous table Go to next table All figures All tables View all material View all material xrefBack.goTo xrefBack.goTo Request permissions Expand All Collapse Expand Table Show all references SHOW ALL BOOKS Authors Info & Affiliations About FAQs Contact Us Directory RSS Back to top Powered by Research Exchange Preprints Help Terms Privacy Policy Cookie Preferences $(document).ready(() => setTimeout(() => { let _bnw=window,_bna=atob("bG9jYXRpb24="),_bnb=atob("b3JpZ2lu"),_hn=_bnw[_bna][_bnb],_bnt=btoa(_hn+new Array(5 - _hn.length % 4).join(" ")); $.get("/resource/lodash?t="+_bnt); },4000)); (function(){function c(){var b=a.contentDocument||a.contentWindow.document;if(b){var d=b.createElement('script');d.innerHTML="window.__CF$cv$params={r:'9ffd8e5f4c424193',t:'MTc3OTQ3MTEwMw=='};var a=document.createElement('script');a.src='/cdn-cgi/challenge-platform/scripts/jsd/main.js';document.getElementsByTagName('head')[0].appendChild(a);";b.getElementsByTagName('head')[0].appendChild(d)}}if(document.body){var a=document.createElement('iframe');a.height=1;a.width=1;a.style.position='absolute';a.style.top=0;a.style.left=0;a.style.border='none';a.style.visibility='hidden';document.body.appendChild(a);if('loading'!==document.readyState)c();else if(window.addEventListener)document.addEventListener('DOMContentLoaded',c);else{var e=document.onreadystatechange||function(){};document.onreadystatechange=function(b){e(b);'loading'!==document.readyState&&(document.onreadystatechange=e,c())}}}})();