Back to Windows Problem Solver home
MPEG-4, also known as ISO/IEC 14496, should have finalized its standardization process in 1999. It reached a major milestone, called "Committee Draft", which means that the MPEG consortium had decided exactly what technical abilities will and will not be included in MPEG-4, and has a very robust draft with demonstrated software for all of these capabilities.
MPEG-4 brings together the worlds of "natural" (recorded) and "synthetic" (synthesized) coding of audio and video. The MPEG-4 Audio standard is built from a toolbox approach; several interconnected forms of audio coding are standardized which together allow a wide range of functionality and flexibility. Here is a list of the audio coders which are part of MPEG-4:
AAC: The Advanced Audio Coding standard, originally created as an extension to MPEG-2, will allow high-quality audio coding. It has been shown in extensive psychoacoustic testing to provide "transparent coding" according to the EBU definition (that is, you can't tell the decoded version from the original) at 64 kbit/sec/channel. It provides better quality at 64 kbit/sec/channel than MP3 does at 128 kbit/sec/channel.
CELP: A codebook-excited linear prediction scheme optimized for telephone- quality transmission of speech in the range 8-32 kbps.
Parametric: A novel "harmonic vector + noise" method which allows lossy but extremely low-bitrate coding of wideband sounds down to 2 kbps/sec/ channel.
Structured Audio: A very powerful downloadable synthesis method which allows producers to describe new synthesis methods as part of the bitstream; the receiver implements a reconfigurable synthesis engine and synthesizes the sound on-the-fly as the instructions are received.
Text-to-Speech: An interface to standalone TTS systems is also provided, so that synthetic speech can be synchronized in multimedia presentations. No "method" of creating synthetic speech is standardized by MPEG.
Also, a tool called Audio BIFS (Binary Format for Scene Description) is standardized in MPEG-4 which allows the synchronization and mixing of sound from multiple coders, and the application of synthetic effects in interactive multimedia scenes. For example, a single MP4 presentation might contain CELP-coded speech, instructions for creating synthetic music, an algorithm for a synthetic reverb to be applied to the speech, and instructions for synchronizing and mixing all of this sound together. Of course it can also be synchronized with visual presentations if desired.
If you want to read about all of these techniques in excruciating detail, the Committee Draft standard is publically available; please check out http://www.cselt.it/mpeg.
There is no "MPEG-1 Layer IV" or "MPEG-3" standard.
Best regards to all,
Eric Scheirer
Editor, MPEG-4 Audio Committee Draft Chair,
MPEG-4 Structured Audio Ad-Hoc group
eds@media.mit.edu
http://sound.media.mit.edu/~eds/mpeg4