The present disclosure relates to a field of intelligent communications, and discloses a speech communication system and method with human-machine coordination, which resolve a problem of bad client experience because great differences occur after a switchover in a call through a prior human-machine coordination and time of a client is wasted. Key points of the technical solutions of the present disclosure include a communications module; an answering and calling-out module, allocating a human agent; a speech intention understanding module for understanding an intention in a call process; a voice cloning module, synthesizing, from a reply text, voice of a corresponding attendant; a human-machine interaction module, to communicate with a client by a speech robot; an intervention prompting module, monitoring call content to obtain an intervention willing probability, and prompting the human agent according to a predefined rule; and a manual intervention module, through which the client is replied by a person, and the manual intervention module displays a call intention and prompts the call during the call. In this way, during the call, a seamless switchover may be performed between the human agent and the speech robot. After the manual intervention, the attendant can quickly understand the call intention, to ensure good call experience of the client.