A method for outputting an audio signal corresponding to scene orientation information is disclosed. The method includes receiving an audio signal interacting with a video, generating a decoded audio signal, object metadata, and scene orientation information through decoding, receiving external control information, and generating modified object metadata suitable for a playback environment by modifying the object metadata based on the received external control information, rendering the decoded audio signal using the modified object metadata, and modifying the rendered audio signal according to the scene orientation information.