Travel speed, is an important parameter for evaluating the operating efficiency of traffic networks, assessing the performance of traffic management strategies, developing real-time vehicle route guidance systems and traffic assignment models. The results of the speed dealy studies are the basic data of the traffic planning projects. Unfortunately, these studies are very expensive, consuming too much time and needing qualified experts. In this study, some travel speed estimation models are presented to predict the average travel speed for developing countries to save money and time. Mixed traffic conditions, comprising of both motorized and non-motorized vehicles on the same carriageway, are significantly different from those in developed countries. Considering this, the data collected from the speed-delay studies in Konya, a city of Turkey, has been used for modeling the traffic flow speed. The measured speeds are classified as road type of links and through a specified alignment. It has been found that the average traffic flow speed is dependent on the traffic rate, number of bicycle in traffic composition, proportion of the people crossing the street, on-street parking quantity and the number of both signalized and non-signalized intersections. The equations have also been regressed by using the generalized least squares method, in order to be able to estimate the average traffic flow speeds from the influencing variables.