Some embodiments provide a method performed by a digital slate. The method receives input of a first set of metadata for a particular video clip from a user. The method receives a second set of metadata for the particular video clip from a camera as the camera films the particular video clip. The method embeds the first and second sets of metadata in a data storage structure. The method transmits the data structure for later use in matching the first and second sets of metadata with the particular video clip. Some embodiments display at least a portion of the stored metadata about the particular video clip in front of a camera at the start of the particular video clip.