Abstract: In the preferred embodiment, a student is able to talk to a pre-recorded actor displayed on a workstation/computer and have the pre-recorded actor respond directly and precisely to the things that the learner is saying, resulting in a fluid and challenging practice conversation without the use of artificial intelligence. A second user (partner) observes the first user (student) either directly or via videoconference or other means and monitors his interaction with the simulated user/actor in the pre-recorded video displayed on the first user's computer screen. The partner has access to an interface that allows her to direct the simulated user/actor by interpreting the first user's communications and selecting among presented options. These options direct the simulated user/actor to respond to the first user's communication in a convincing and helpful manner.