* * * * * * * * * * * * * * * * * * * * * * * *
LINGUIST List logo Eastern Michigan University Wayne State University *
* People & Organizations * Jobs * Calls & Conferences * Publications * Language Resources * Text & Computer Tools * Teaching & Learning * Mailing Lists * Search *
* *


LINGUIST List 23.3643

Fri Aug 31 2012

Diss: Phonetics: Jekale: 'An acoustic analysis of Amharic vowels, plosives and ejectives'

Editor for this issue: Lili Xia <lxialinguistlist.org>

Date: 30-Aug-2012
From: Derib Jekale <derib.adogmail.com>
Subject: An acoustic analysis of Amharic vowels, plosives and ejectives
E-mail this message to a friend

Institution: Addis Ababa Univiersity
Program: Experimental Phonetics
Dissertation Status: Completed
Degree Date: 2011

Author: Derib Ado Jekale

Dissertation Title: An acoustic analysis of Amharic vowels, plosives and ejectives

Linguistic Field(s): Phonetics

Dissertation Director:
Moges Yigezu
Klaus Wedekind

Dissertation Abstract:

This research set out to study the vowels, and pulmonic and ejective stops of Amharic, a South Ethio-Semitic language spoken in Ethiopia. Data from four males and four females was recorded at the Phonetics lab established at the Akaki Campus of Addis Ababa University using the Kay CSL 4400 multisignal acquisition and analysis device.

The F0, F1, F2, F3 values of Amharic vowels in /tVt/ context were computed and presented. The F1-F2 plot of Amharic vowels showed that the so called 'high central vowel' was not really high, but high mid, and the so called 'mid central vowel' was found to be not mid, but low mid, close to the cardinal vowel [ɜ]. Thus a new four-height vowel system has been proposed based on the results of this study. Amharic vowels were found longer before voiced stops than before voiceless stops, and they were longer before singletons than before geminates conforming to attested patterns in other languages. A discriminant analysis showed that F1 and F2 alone could be used to classify with 86.7% accuracy, whereas the addition of F0 and F3 values in the discriminant analysis increased it to 90% correct classification for all subjects collapsed across genders. Using Normalized values the classification results rose to 94%.

The acoustic analysis of stops showed that total burst duration and relative intensity came out strong in classifying stops by place of articulation. Bilabials had the highest classification results followed by velars and alveolars had the least classification results. VOT and spectral mean were the most important acoustic cues that classified voiced stops with a correct classification result of 94.9% and voiceless stops up to 89%. VOT alone was able to classify 81.4% of the stops correctly. Jitter perturbation, spectral mean and voicing lag came out as relatively stronger acoustic cues that identified airstream. Nevertheless, classification using the numerical results of jitter perturbation, spectral mean and voicing lag did not produce expected results though it has been found that Amharic ejectives result in creaky phonation on the following vowel. Amharic ejectives have been found to have more characteristics of slack ejectives than of stiff ejectives.




Read more issues|LINGUIST home page|Top of issue



Page Updated: 31-Aug-2012

Supported in part by the National Science Foundation       About LINGUIST    |   Contact Us       ILIT Logo
While the LINGUIST List makes every effort to ensure the linguistic relevance of sites listed on its pages, it cannot vouch for their contents.