Recent work at the Sussex Humanities Lab, a digital humanities research program at the University of Sussex, has sought to address an identified gap in the provision and use of audio feature analysis for spoken word collections. Traditionally, oral history methodologies and practices have placed emphasis on working with transcribed textual surrogates, rather than the digital audio files created during the interview process. This provides a pragmatic access to the basic semantic content, but obviates access to other potentially meaningful aural information; our work addresses the potential for methods to explore this extra-semantic information, by working with the audio directly. Audio analysis tools, such as those developed within the established field of Music Information Retrieval (MIR), provide this opportunity. This paper describes the application of audio analysis techniques and methods to spoken word collections. We demonstrate an approach using freely available audio and data analysis tools, which have been explored and evaluated in two workshops. We hope to inspire new forms of content analysis which complement semantic analysis with investigation into the more nuanced properties carried in audio signals.
MethodologyFor the past decade, 3D archaeological visualisations have mostly been representing photo-realistic reconstructions of ancient monuments. While these can be constructive in a museum or tourist context, the archaeological community has long stressed the need for reconstructions showing where the actual remains end and the assumptions begin. Recent attempts to implement the latter approach are either limited to found/not found scenarios or marking of uncertain areas without any justification to the choice of colour/hue degradation etc. As a result, there is no system to represent the uncertainty involved in visualising archaeological data. The archaeologist interprets a site based on a limited amount of material remains and uses comparative evidence from other sites, written references, as well as speculation in order to create a reconstruction. Due to this varied range of data, he/she may have different levels of certainty on some areas of the reconstruction than others. If we are able to observe this uncertainty on the visualisation itself, it would provide us with a whole new range of uses for archaeological models, such as learning about archaeological hypotheses, comparing uncertainties across different models and highlighting cases where further research may be required.Previous research has indicated that a fuzzy logic approach can represent the uncertainty of a complete archaeological reconstruction. However, fuzzy logic is specifically designed to deal with imprecision of facts -such as "the degree to which a fragmented tile belongs to the family of Roman tiles," rather than describing an archaeologist's interpretation confidence. Our approach uses possibility theory [Dubois and Prade 1988], an extension of fuzzy sets and fuzzy logic, to represent uncertainty in visualisation; it is a mathematical theory for dealing with uncertainty and an alternative to probability theory. While the latter works best with precise but varied knowledge, possibility theory does not expect such precise information -but we do hope for the greatest possible coherence from experts. Possibility theory has been used to describe human uncertainty in medical case studies with success and one of its strengths is the ability to provide both ordinal and numerical answers, whereas probability allows only for numerical. To briefly summarise its basic functions, let X be a variable for which we do not have full knowledge i.e. the shape of a capital missing from a Greek column, and Ω x the set of values X can take (e.g. Ionic, Doric). A possibility distribution π x (ω) → [0, 1] describes the extent to which it is possible that the actual value of X be Doric (ω). Let A be an event subset of Ω, so the possibility measure that A is correct is Π(A) = sup ω∈A Π x (ω). Π(A) expresses the level of possibility that the capital is one of the set associated with A. Possibility has also a complementary measure, necessity, which represents the possibility of the contrary event: N(A) = 1 − Π(¬A) Combination of the two measures aids in statem...
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.