A device configured to receive a video request that includes animation instructions for a video scene. The animation instructions identify one or more animations associated with the video scene. The device is further configured to identify a first animation from the one or more animations associated with the video scene and to determine that the first animation is configured for text-to-audio. The device is further configured to identify text associated with the first animation and to convert the text associated with the first animation into an audio sample. The device is further configured to associate the audio sample with an animation identifier for the first animation in an audio sample buffer. The device is further configured to associate a timestamp with a source scene identifier for the video scene and the animation identifier for the first animation in the video timing map.