The adoption of smart phones, the increased access to mobile broadband networks and the availability of public cloud infrastructures are aligning to the next generation of truly ubiquitous multimedia services, known as Cloud Mobile Media (CMM) services offering mobile video. Nevertheless, due to an inherit higher and variable end to end delay mainly as a result of the virtualization process, new challenges appear. One challenge is given by live video streaming applications when trying to keep a good Quality of Experience of the delivered video, measured in terms of a subjective video quality metric, named Mean Opinion Score (MOS). Our goal is to estimate and predict this subjective metric in a holistic manner using different estimation techniques, such as Artificial Neural Networks, Factor Analysis and Multinomial Linear Regression, with Full Reference and Non Reference approaches. For this, we have analyzed and measured different variables related to Quality of Service, bit stream and basic video quality metrics, throughout the CMM infrastructure. With these variables, we apply the mentioned techniques which allows us to estimate MOS of the delivered video in a robust and reliable way, achieving an average estimation error between 0.46 and 15.94% depending on the technique used. The real MOS has been evaluated through surveys. Finally, we compare the accuracy of the estimated MOS against well known publicly available video quality algorithms, following the recommendations given by Video Quality Experts Group.