Atlas.txt: exploring linguistic grounding techniques for communicating spatial information to blind users

Abstract

This paper describes exploratory research into automatically describing geo-referenced information to blind people. The goal is to produce texts giving an overview of the spatial layout, and a central concern of such texts is that they employ an appropriate linguistic reference frame which enables blind hearers to ground the information. The research presented in this paper was based on two hypotheses: (1) directly perceivable reference frames are easier to ground, and (2) spatial descriptions drawn from composite reference frame systems composed of more than one reference frame are easier to ground. An experiment exploring text comprehension on a range of texts employing different reference frame systems is presented. The main results indicate that the second hypothesis is supported. A prototype of a natural language generation system which generates texts describing geo-referenced information from data is described.Keywords blind users · geo-referenced spatial information · natural language generation· reference frame preference 1

The impact of culture and recipient perspective on direction giving in the service of wayfinding

“…In a recent study, this idea was put to the test by finding out the most effective way to describe census data to people who are blind (Thomas, Sripada, & Noordzij, 2012). Census data often contain information about the frequency distribution of demographic variables (i.e., incidence of crime in different regions in a country).…”

Making Visual Maps Accessible to the Blind

“…audio and haptic), the project Atlas.txt [21] investigated how geo-referenced information (often conveyed via shaded thematic choropleth maps) can be communicated to blind users; the project developed a prototype data-to-text Natural Language Generation (NLG) system that produces textual summaries of UK 2001 Census data.…”

Modeling Spatial Knowledge from Verbal Descriptions

“…Navigation can be modeled as a sequence of locations and actions from an initial location to a target location. It has been shown that verbal descriptions are well suited to the description of human navigations [7,8,14]. Several recent studies have attempted to model navigation knowledge from human verbal descriptions [12,16].…”

“…However, and despite the fact that humans are effective to process and transmit such knowledge, there is still a conceptual gap between these descriptions and map representations. Other approaches in indoor or outdoor environments have been suggested to model verbal routes descriptions [8,14].…”