Spectrographic Analysis of Dagbani Vowels Fusheini Abdul-rahman Acknowlagement

The study presents an acoustic analysis of 15 stressed short and long vowels of Dagbani with detailed formant frequency and duration characteristics of the vowels produced in the open monosyllabic (CV) phonetic context by 15 male and 15 female speakers of Nanuni, Nayahili, and Tomosili dialects of Dagbani. The aim of the study was to provide an acoustic evidence of vowel quality in Dagbani through comparison of the acoustic characteristics of the vowels produced by speakers of the three dialects of Dagbani. The voice recording of 30 speakers made up of five males and five females for each dialect was carried out. The speech data was analyzed on the Kay Elemetrics Computerized Speech Lab (CSL model 4500) software to provide frequency data and duration values for the vowels. Values for the lowest two formants (F1 and F2) for speakers’ vowels were taken at the middle and the durations measured from the start to the end of the study state of each vowel. Both frequency and statistical results indicate that no significant differences exist between the dialects of Dagbani in the CV environment. The results also reveal that Nanuni which for the first time has been studied as a dialect of Dagbani is not different from Nayahili and Tomosili on which most linguist have often based their studies as dialects of Dagbani. The position of [e] and [ɛ] across the dialects as well as gender suggests that Dagbani does not have [e] and [ɛ] distinction, at least, in CV context. Speakers did not also distinguish between the production of [i] and [ ]. Further more, the results show that [a] is more of a low back vowel than a mid one. It also reveals that the sound in Dagbani labeled [ɨ] is the mid central vowel [ə] and not the mid high vowel [ɨ]. The results also confirm that two degrees of vowel length exist in Dagbani; the short vowels being about 62.4% of their longer counterparts. CHAPTER ONE GENERAL INTRODUCTION