Abstract

To describe the affects related to the perception of sound environments, the ISO 12913 standard proposes the use of 8 verbal scales (pleasant, chaotic, vibrant, uneventful, calm, annoying, eventful and monotonous) inspired by the theory of emotions proposed by Russell et al., and the Swedish Sounsdcape Quality Protocol. In the context of the international development of this standard, the issue of translating these terms into each language is currently being studied by about 20 research teams around the world in the framework of the “Soundscape Attribute Translation Project”. As a counterpoint to this approach, for languages that will not be considered in the project or for illiterate populations (e.g. children), this study proposes to evaluate the relevance of using facial emojis to translate the verbal attributes. This work proposes a replication of the protocol set out in the “Soundscape Attribute Translation Project” by presenting facial emojis instead of the verbal attributes. A descriptive analysis of the results through a principal component analysis suggests that the variability in the description of emotions with non-verbal attributes can be decomposed on two principal axes (valence and arousal), using verbal attributes and that among 8 verbal attributes, 5 could be replaced by emojis in a direct way. Within the framework of ISO/TS 12913–3 and by transforming the attributes on the two main axes ISOPleasant and ISOEventful, a comparative analysis was also carried out with an experiment following the same protocol but using French verbal attributes. The latter shows a strong correlation between the positioning of the audio recordings on the ISOPleasant (r = 0.95, p < 0.001) and ISOEventful (r = 0.77, p < 0.001) axes. The results are encouraging and the use of facial emojis could potentially be proposed within the ISO standard, although a future extension of this study should be conducted by comparison with other languages/cultures and their acceptance by illiterate populations to confirm the relevance of using facial emojis to translate soundscape descriptors.