Cross-modal codification of images with auditory stimuli: a language for the visually impaired

TL;DR

This paper presents a novel method converting images into sounds to aid visually impaired individuals, demonstrating high discrimination accuracy and potential for object recognition through auditory cues.

Contribution

It introduces an original algorithm for image-to-sound conversion and demonstrates its effectiveness in object discrimination and recognition tasks.

Findings

Subjects discriminated sounds with 95% accuracy for different shapes.

Subjects recognized target images among 16 options better than chance in 90% of cases.

Recognition rate depended on image characteristics.

Abstract

In this study we describe a methodology to realize visual images cognition in the broader sense, by a cross-modal stimulation through the auditory channel. An original algorithm of conversion from bi-dimensional images to sounds has been established and tested on several subjects. Our results show that subjects where able to discriminate with a precision of 95\% different sounds corresponding to different test geometric shapes. Moreover, after brief learning sessions on simple images, subjects where able to recognize among a group of 16 complex and never-trained images a single target by hearing its acoustical counterpart. Rate of recognition was found to depend on image characteristics, in 90% of the cases, subjects did better than choosing at random. This study contribute to the understanding of cross-modal perception and help for the realization of systems that use acoustical signals to help visually impaired persons to recognize objects and improve navigation