MA / Department of Communication and Applied Language Studies / Orthographic symbols of languages represent specific speech sounds with their specific phonetic qualities. This research aimed at analysing orthographic representations of murmured speech sounds in the Xitsonga orthography. The qualitative approach was employed to examine the production of murmured speech sounds and their orthographic representations in Xitsonga. The analysis was based on data collected from interviews with speakers of the Xitsonga language, Xitsonga educators and lecturers who are themselves Xitsonga-speaking, Xitsonga authors and Xitsonga subject specialists, and also from existing literature by Xitsonga authors and language scholars. The study made recommendations that will help the development and revitalisation of the Xitsonga language and also benefit the speakers of the language as well as scholars and linguists.

comprises 50 occurrences of presenter's IPS and 50 occurrences of guests' IPS. - - - - ' presenter's IPS react in 98 % of cases to (78 % of presenter's IPS occur in thematic , on the other hand 52 % of guests' IPS occur in thematic (while presenter's IPS adjacency pair, guests' IPS in most -

Indiana University-Purdue University Indianapolis (IUPUI) / To access interactive systems, blind and visually impaired users can leverage their auditory senses by using non-speech sounds. The current structure of non-speech sounds, however, is geared toward conveying user interface operations (e.g., opening a file) rather than large theme-based information (e.g., a history passage) and, thus, is ill-suited to signify the complex meanings of primary learning material (e.g., books and websites). In order to address this problem, this dissertation introduces audemes, a new category of non-speech sounds, whose semiotic structure and flexibility open new horizons for facilitating the education of blind and visually impaired students. An experiment with 21 students from the Indiana School for the Blind and Visually Impaired (ISBVI) supports the hypothesis that audemes increase the retention of theme-based information. By acting as memory catalysts, audemes can play an important role in enhancing the aural interaction and navigation in future sound-based user interfaces. For this dissertation, I designed an Acoustic EDutainment INterface (AEDIN) that integrates audemes as a way by which to vividly anticipate text-to-speech theme-based information and, thus, act as innovative aural covers. The results of two iterative usability evaluations with total of 20 blind and visually impaired participants showed that AEDIN is a highly usable and enjoyable acoustic interface. Yet, designing well-formed audemes remains an ad hoc process because audeme creators can only rely on their intuition to generate meaningful and memorable sounds. In order to address this problem, this dissertation presents three experiments, each with 10 blind and visually impaired participants. The goal was to examine the optimal combination of audeme attributes, which can be used to facilitate accurate recognitions of audeme meanings. This work led to the creation of seven basic guidelines that can be used to design well-formed audemes. An interactive application tool (ASCOLTA: Advanced Support and Creation-Oriented Library Tool for Audemes) operationalized these guidelines to support individuals without an audio background in designing well-formed audemes. An informal evaluation conducted with three teachers from the ISBVI, supports the hypothesis that ASCOLTA is a useful tool by which to facilitate the integration of audemes into the teaching environment.