In an approach to determine a pre-recorded video for use in place of a video feed of a user in a video conference, a computer receives one or more pre-recorded videos of a user in which the pre-recorded videos include at least one visual representation of an emotional state. The computer determines a meeting tone of at least one meeting discussion in the video conference and determines a first video of the one or more pre-recorded videos of the user that corresponds to the meeting tone. The computer substitutes the first video into the video conference in place of a video feed in the video conference.