This research describes syllables of Kimunasukuma, a dialect of Kisukuma language spoken in Lake Victoria zone. The objectives of this study were; to identify the syllable structure of Kimunasukuma dialect, to describe the relationship between underlying representation and surface representation, and to explain the phonological processes undergone by syllables in Kimunasukuma dialect. The study used Kimunasukuma native speakers from Nyanguge and Kisesa villages which are found in Magu district of Mwanza region. In data presentation and analysis the study uses a generative CV-phonology model of syllable structure as postulated by Clements and Keyser (1983). In this theory Clements and Keyser requires the theory of the syllable to perform three tasks, state universal principles governing syllable structure; (b) state syllable structure typology, and to state language-specific rules governing syllable structure. The theory presents data by describing the syllable into three-tiered structure consisting of a syllable node, a CV-tier and segmental tier. The researcher used unstructured interview, questionnaires and documentation in collecting data on the study. The study presents various syllable structure of Kimunasukuma dialect; it also presents six phonological processes which affects sound. However the study discusses the relationship between underlying and surface representation where it is observed that underlying representation relates with surface representation in the language of study. The study recommends the way forward on topics to be conducted on Kimunasukuma dialect in order to make the language well researched.