Streets form a crucial component of urban infrastructure, serving as vibrant conduits that interlink people, local history, and the community within a city. Despite the recognized significance of street space quality in enhancing the liveliness of urban areas, there is a notable scarcity of studies specifically dedicated to assessing the quality of these spaces. However, with the availability of online datasets and advancements in machine learning, it has become feasible to use images to evaluate the visual quality of streets. Against this backdrop, this study uses Google Street View images to assess the visual quality of street spaces in Singapore through machine learning modelling. The results demonstrate that the Support Vector Regression (SVR) model can effectively quantify the visual quality of a street space. Furthermore, the SVR model's results align with accompanying survey results, indicating that machine learning approaches can comprehend the landscape attributes in urban streetscapes and offer a perspective to guide landscape aesthetic assessments in the built environment.