Intelligent voice user interfaces (UIs) are increasingly accepted by users to replace traditional ones to operate intelligent personal assistant devices. This study developed a model to explicate the key factors influencing users’ voluntary switching behavior between traditional UIs and intelligent voice ones from push-pull-mooring framework perspective, which considers the push effect from two dimensions of perceived inconvenience, the pull effect from three factors based on perceived value, and the mooring effect from switching costs. A mixed-methods approach is adopted to explore pull factors (qualitative) and validate model (quantitative). The model was validated based on 259 respondents using both traditional UI and intelligent voice one of smart home devices from China. The results indicated that all push (extrinsic and intrinsic inconvenience), pull (interface adaptivity, playfulness, and social presence) and mooring (switching costs) factors significantly affect switching intention and then affect actual behavior. The implications for research and practice are discussed.