LINGUIST List 17.806
Thu Mar 16 2006
Diss: Phonetics: Mády: 'Akustische, artikulatorische...'
Editor for this issue: Takako Matsui
<takolinguistlist.org>
Directory
1. Katalin
Mády,
Akustische, artikulatorische und perzeptive Parameter in der Konsonantenproduktion nach Zungenteilresektion
Message 1: Akustische, artikulatorische und perzeptive Parameter in der Konsonantenproduktion nach Zungenteilresektion
Date: 16-Mar-2006
From: Katalin Mády <madybtk.ppke.hu>
Subject: Akustische, artikulatorische und perzeptive Parameter in der Konsonantenproduktion nach Zungenteilresektion
Institution: University of Munich
Program: Department of Phonetics and Speech Communication
Dissertation Status: Completed
Degree Date: 2004
Author: Katalin Mády
Dissertation Title: Akustische, artikulatorische und perzeptive Parameter in der Konsonantenproduktion nach Zungenteilresektion
Dissertation URL: http://edoc.ub.uni-muenchen.de/archive/00003060/01/Mady_Katalin.pdf
Linguistic Field(s):
Phonetics
Dissertation Director:
Hans-Henning Horch
Hans G. Tillmann
Dissertation Abstract:
In the thesis, acquired organic speech disorders due to oral cavity tumoursare investigated under acoustic, articulatory and perceptual aspects.
Parts of the investigation are new in this field and have an explorativecharacter, thus it was chosen to keep the number of subjects and the amountof analysed consonants relatively small (8 males, 4 sounds) and theanalysis as detailed as possible.
Articulatory recordings were done by dynamic MRI in the midsagittal plane(8 images/sec), and the analysis was based on the distance between tongueand palate. The acoustic analysis relied on separate recordings andincluded parameters which are relevant for the segment characterisation:duration, relative intensity, and spectral analysis. The study alsoinvolved an auditory evaluation, an acoustic investigation and ameasurement of the overall speech ability by an intelligibility and by aperception test.
The investigation of the postoperative realisations of the sibilants /s/and /S/ led to a somewhat surprising result. It is generally thought that/S/ requires a greater articulatory effort than /s/, what results in alater acquisition during first language acquisition and in a less universalcharacter among different languages. However, both the acoustic andarticulatory analysis revealed, that impaired sibilant production almostalways lead to a neutralisation, where the acoustic and articulatoryparameters of /s/ and /S/ overlapped widely and showed the characteristicsof /S/. It seems that the crucial distinctive factor for /s/ and /S/ israther the tongue shape than the point of articulation. These findings makeit necessary to re-examine former hypotheses on the differentiation betweenalveolar and postalveolar sibilants that are based on a sublingual cavity(Perkell, Boyce and Stevens, 1979) or on a secondary obstacle (Shadle, 1990).
The most severe production difficulties were connected to the alveolarlateral /l/. The apical production, that is characteristic for German,could be achieved only by one of 8 subjects. This could be explained by thefact that the mobility of the tongue tip was restricted in all othersubjects. Another explanation would be that the function of genioglossusmuscle, that plays a role in the lowering of the tongue body, was actuallyimpaired in all but one subject. Theories regarding the relation of F1, F2,and the F2-F3 distance for apical vs. laminal /l/ production cannot beverified on our data, as the effects that lead to a shift in F1 and F2 areoften contrary regarding the point of articulation and the tongue bodyposition.
The velar fricative /x/ did not cause a problem to any of the subjects. Thenormal production of this sound shows a large articulatory variation (velarand uvular) and does not require exact tongue movements. The choice of thevelar or uvular allophone seems to vary not only among persons, but alsointra-personally (among recording settings following directly after eachother): for one subject, the MRI showed a velar preoperative articulation,while the acoustic characteristics referred to an uvular production. It wasobserved that the number of uvular realisations was somewhat higherpostoperatively. This might result from the fact that the uvular soundrequires a smaller degree of tongue body elevation than the velar one.
The intelligibility test revealed that the overall perceptual evaluation isnot only influenced by the size and localisation of the resection, but alsoa higher speech rate can lead to lower intelligibility scores.
|