This document describes the desired joint behaviour of MPEG-4 Systems (MPEG-4 File Format) and MPEG-4 Audio codecs. It is desired that MPEG-4 Audio encoders and decoders permit finite length signals to be encoded to a file (particularly MPEG-4 files) and decoded again to obtain the identical signal, subject to codec distortions. This enables the use of audio in systems implementations (particularly MPEG-4 Systems), perhaps with other media such as video, in a deterministic fashion. Most importantly, the decoded signal has nothing “extra” at the beginning or “missing” at the end.

This permits:

a)       an exact "round trip" from raw audio to encoded file back to raw audio (excepting encoding artefacts);

b)       predictable synchronization between audio and other media such as video;

c)        correct behaviour when performing random access as well as when starting at the beginning of a stream;

d)       identical behaviour when edits are applied in the raw domain and the encoded domain (excepting encoding artefacts).

It is also expected that there be predictable interoperability between encoders (as represented by files) and decoders. There are two kinds of audio "offsets" (or "delay" in the context of transmission): those that are result from the encoding process, and those that are result from the decoding process. This document is primarily concerned with the latter.

These issues are resolved by the following:

—     The handling of composition time stamps for audio composition units is specified. Special care is taken in the case of compressed data, like HE-AAC coded audio, that can be decoded in a backward compatible fashion as well as in an enhanced fashion.

—     Examples are given that show how a finite length signals can be encoded to an MPEG-4 file and decoded again to obtain the identical signal, excepting codec distortions. Most importantly, the decoded signal has nothing “extra” at the beginning or “missing” at the end.