Developed by O＇Beirne and Trounson (Trounson, 2012), the UC Auditory-Visual Matrix Sentence Test (UCAMST) is an auditory-visual speech test in NZ English where sentences are assembled from 50 words arranged into 5 columns (name, verb, quantity, adjective, object). Generation of sentence materials involved cutting and re-assembling 100 naturally spoken ‟original” sentences to create a large repertoire of 100,000 unique ‟synthesised” sentences. The process of synthesising sentences from video fragments resulted in occasional artifactual image jerks (‟judders”)‒quantified by an unusually large change in the ‟pixel difference value” of consecutive frames‒at the edited transitions between video fragments. To preserve the naturalness of materials, Study 1 aimed to select transitions with the least ‟noticeable” judders. Normal-hearing participants (n = 18) assigned a 10-point noticeability rating score to 100 sentences comprising unedited ‟no judder” sentences (n = 28), and ‟synthesised” sentences (n = 72) that varied in the severity (i.e. pixel difference value), number, and position of judders. The judders were found to be significantly noticeable compared to no judder controls, and based on mean rating score, 2,494 sentences with ‟minimal noticeable judder” were included in the auditory-visual UCAMST. Follow-on work should establish equivalent lists using these sentences. The average pixel difference value was found to be a significant predictor of rating score, therefore may be used as a guide in future development of auditory-visual speech tests assembled from video fragments. The aim of Study 2 was to normalise the auditory-alone UCAMST to make each audio fragment equally intelligible in noise. In Part I, individuals with normal hearing (n = 17) assessed 400 sentences containing each file fragment presented at four different SNRs (-18.5, -15, -11.5, and -8 dB) in both constant speech-shaped noise (n = 9) and six-talker babble (n = 8). An intelligibility function was fitted to word-specific data, and the midpoint (Lmid, intelligibility at 50%) of each function was adjusted to equal the mean pre-normalisation midpoint across fragments. In Part II, 30 lists of 20 sentences were generated with relatively homogeneous frequency of matrix word use. The predicted parameters in constant noise (Lmid = 14.0 dB SNR; slope = 13.9%/dB ± 0.0%/dB) are comparable with published equivalents. The babble noise condition was, conversely, less sensitive (Lmid = 14.9 dB SNR; slope = 10.3%/dB ± 0.1%/dB), possibly due to a smaller sample size (n = 8). Overall, this research constituted an important first step in establishing the UCAMST as a reliable measure of speech recognition; follow-on work will validate the normalisation procedure carried out in this project.

The ability to understand speech in noise has a profound impact on everyday communication, but cannot be predicted on the basis of puretone thresholds and/or performance on tests of speech in quiet. The aim of this thesis was to develop an adaptive speech in noise test based on the Hearing in Noise Test (HINT) that would be reliable and valid for speakers of New Zealand English (NZE). The methodology used followed the standard procedures for developing the HINT in a new language. Five hundred sentences of 5-7 syllables were collected from New Zealand children’s books and recorded by a native NZE speaker. Nine normal-hearing native NZE speakers aged 18-50 listened to three sets of 50 sentences at -2, -4 and -7 dB signal-to-noise ratios (SNR) in order to establish a performance-intensity (PI) function for these sentences. Three groups of 10 participants were scored on their performance on the sentences in 65 dBA speech-weighted noise at varying SNR. After each round of testing with a new group of participants, the SNR of each sentence was adjusted in order to get closer to 70% intelligibility for all sentences. Sentences that were too easy or difficult or did not respond to adjustments were discarded. Once the remaining 240 sentences were of approximately equal intelligibility, 24 phonemically matched lists of 10 sentences were formed and tested on 12 participants using the adaptive HINT software. The overall mean threshold was calculated as -6 dB, s.d=1.1 dB. The lists were combined to form 12 lists of 20 sentences which would become the NZHINT. Time delays meant that the collection of normative data could not be completed.

Background: The difficulties for people to understand and perceive speech in second languages in the presence of background noise is well documented in the literature. Similar difficulties have been reported in bilinguals. Consequently, the speech-in-noise tests can be used as a diagnostic tool to assess an individual's speech perception in noisy environments and plan effective hearing rehabilitation but it has less precision in the case of people whose first language is different from the tested language. On the other hand, the newly developed Situated Phoneme Test (SiP) might be a promising diagnostic method for evaluating speech perception in noise. The advantage is that the test person does not have to repeat the words as in the TiB test - something that makes it difficult for people with a different first language - and therefore the source of error of subjective assessment by the audiologist is eliminated. Aim: The aim of the present work is to evaluate the effectiveness of the SiP-test, for a group of people with Arabic as their first language, by examining the results of the SiP-test and the results of the conventional TiB test in relation to expected results for those who have Swedish as first language. A secondary aim is to evaluate people's experiences of using the TiB -test and the SiP-test. Materials and methods: Speech perception in noise was investigated in 32 normal-hearing individuals using the TiB test and the SiP-test. Common to all participants was that they had Arabic as their first language, moved to Sweden after the age of 8 and had good knowledge of the Swedish language according to the results of the language test they completed. The results of the TiB -test and the SIP-test were analyzed and compared with the expected values for people whose first language is Swedish. In addition, two diagnostic methods were evaluated by the participants using a questionnaire, specially designed for this purpose. Results: Overall, the SiP-test results for this group of people with Arabic as first language were similar or better (M = 83.1, SD = 4.31) than expected (M = 75, SD = 0) for those whose first language is Swedish, while the mean value for the TiB -test results was worse (M = 72.9, SD = 2.06) than expected (M = 84, SD = 0) for native Swedishspeakers. The TiB test results showed that no participant managed to reach the expected values based on people whose first language is Swedish. The individuals reported, with the use of a questionnaire, that the SiP-test was easier to perform (n = 18). Participants mentioned " the repeating of the words" as the more difficult for the TiB-test (n = 32) and "the use of the touch screen" as the easier for the SiP-test (n = 30). Conclusions: The results of the present study showed that SIP-test can be an accurate and reliable diagnostic method for the evaluation of speech perception in noise in non-native speakers, compared to the conventional TiB-test. SIP-test proved to be an easy, safe, realistic and objective diagnostic tool, based on the results and the reports of the participants who completed the test. Further research in individuals with various degrees and types of hearing loss, as well as similar studies in populations having other languages than the Arabic as its first language are needed in order to fully investigate the use and clinical applications of the SIP-test as a valuable part of an integral hearing investigation on the everyday clinical practice. / Bakgrund: Svårigheterna för personer att förstå och uppfatta tal på andraspråket i närvaro av bakgrundsbrus är väldokumenterade i litteraturen. Liknande svårigheter har rapporterats även hos tvåspråkiga. Följaktligen kan tal-i-brus-tester (TiB) användas som ett diagnostiskt verktyg för att bedöma en individs taluppfattning i bullriga miljöer och planera en effektiv hörselrehabilitering men det har mindre precision när det gäller personer som har ett annat förstaspråk än det testade språket. Det nyutvecklade Situated Phoneme Test (SiP) kan eventuellt vara en lovande diagnostisk metod för att utvärdera taluppfattning i buller. Fördelen med SiP-testet är att testpersonen inte behöver upprepa orden som i TiB-testet - något som gör det svårt för personer med ett annat förstaspråk - och därav elimineras felkällan av subjektiv bedömning från audionomen. Syfte: Syftet med denna studie är att utvärdera effektiviteten av SiP-testet, för en grupp av personer med arabiska som förstaspråk, genom att undersöka resultaten av SiP-testet och resultaten av det konventionella TiB-testet i relation till förväntade resultat för dem som har svenska som förstaspråk. Ett sekundärt syfte är att utvärdera personernas erfarenheter av att använda TiB-testet och SiP-testet. Metod och material: Taluppfattningen i brus undersöktes hos 32 normalhörande individer med hjälp av TiB-testet och SiP-testet. Gemensamt för samtliga deltagare var att de hade arabiska som förstaspråk, hade flyttat till Sverige efter 8 års ålder och hade goda kunskaper i det svenska språket enligt resultatet av det språktest de genomförde. Resultaten av TiB- och SiP-testerna analyserades och jämfördes med de förväntade värdena för personer som har svenska som förstaspråk. Dessutom utvärderades två diagnostiska metoderna av deltagarna med hjälp av ett frågeformulär, speciellt utformat för detta ändamål. Resultat: Sammantaget var SiP-testresultaten för denna grupp av personer med arabiska som förstaspråk liknande eller bättre (M = 83.1, SD = 4.31) än förväntat (M = 75, SD = 0) för de som har svenska som förstaspråk, medan medelvärdet för TiB-testresultaten var sämre (M = 72.9, SD = 2.06) än förväntat (M = 84, SD = 0) för de som har svenska som förstaspråk. TiB-testet resultatet visade att ingen deltagare lyckades nå de förväntade värdena baserade på personer som har svenska somförstaspråk. Flest av deltagarna rapporterade, genom frågeformuläret, att SiP-testet var lättare att utföra (n = 18). Deltagarna nämnde ”att upprepa orden" som det svårare för TiB-testet (n = 32) och "att använda pekskärmen" som det lättare för SiP-testet (n = 30). Slutsatser: Resultaten av den föreliggande studien visade att SiP-test kan vara en korrekt och tillförlitlig diagnostisk metod för att utvärdera taluppfattningen för personer med svenska som andraspråk. SiP-test visade sig vara ett enkelt, säkert, realistiskt och objektivt diagnostiskt verktyg, baserat på resultaten och rapporterna från deltagarna som genomförde testet. Ytterligare forskning på individer med olika grader och typer av hörselnedsättning, såväl som liknande studier i populationer som har andra språk än arabiska som sitt förstaspråk behövs för att fullständigt undersöka användningen och kliniska tillämpningar av SIP-testet som en värdefull del av en integrerad hörselundersökning om den kliniska vardagen.