Abstract: In some implementations, a positional audio assignment system is used to improve a user's immersion during content playback within a virtual reality setting. Data representing a video viewable to a user identifying spatial positions assigned to one or more objects within the video is initially obtained. Audio data encoding one or more audio streams corresponding to each of the one or more objects is also obtained. User input data associated with playback of the video is then received. A gaze point of the user based on the received user input data is then determined. The gaze point of the user is then evaluated with respect to the spatial positions assigned to the one or more objects. The audio output provided to the user is then selectively adjusted based on evaluating the gaze point with respect to the spatial positions assigned to the one or more objects.