MPEG Multichannel Audio Encoding

The introduction of DVD as the carrier for the next generation of high quality digital movies has brought about the development of encoding products for both the video information itself, and the accompanying soundtrack. For multi-channel soundtracks, Dolby Digital and / or MPEG2 are the standard audio compression technologies for DVD in PAL/SECAM countries.

Variable Bit Rates

The compression technologies define only the bit rate range boundaries for audio encoding. The encoding bit rate can vary in accordance with the complexity of the sound to be encoded. The variable bit rate used by DVD is a very efficient way of encoding, as each piece of sound is encoded with a bit rate that is sufficient to encode it without any artefacts. Therefore, a simple sound is encoded at a much lower bit rate than a complex piece. In this way, the average bit rate is much lower than that of a fixed bit rate encoder, with absolutely no compromise in audio quality. This is especially applicable to movie soundtracks where sound complexity can vary very widely within short spaces of time.

Multichannel Audio

The separate specialised hardware unit is used to encode multi-channel audio (5.1 channel). The audio consists of up to 5 full bandwidth channels (0 Hz - 20 kHz), plus an optional low frequency enhancement (LFE) channel (0 Hz - 120 Hz). Optionally, two further full bandwidth channels can be handled (7.1) in the MPEG-2 format. As specified in the DVD standard, the encoder provides facilities for generating a stereo or Dolby Surround down-mix in the MPEG-2 and Dolby Digital bit-stream, providing compatibility with all types of decoder.

MPEG2 audio encoding process diagram

MPEG2 Audio Encoding Hardware

First, the audio master must be ‘grabbed’ or captured from the master tape (e.g. Digital Betacam, TASCAM DA-88 or DAT) to hard disk using a software package. This is simply a matter of defining the start and stop times of the soundtrack, and some basic properties of the audio master, then allowing the software to write the data to hard disk.

The captured audio data is fed in to the encoder through the PCI bus. Encoding is an equally straightforward process as grabbing. All that is required is specification of the un-encoded audio file, the output MPEG2 file, and some basic encoding parameters (e.g. bit rate, down-mix type), then the encoding process can be started.

The main operations in encoding are the matrixing and sub-band filtering. The other major task of the encoder is the allocation of the multi-channel parts into the transmission channel, and multiplexing these into the MPEG2 bit-stream. An MPEG2 frame is structured in parts: an stereo compatible part (channels 1+2), with extensions for the 5.1 part (channels 3-5 + LFE). The 7.1 part (channels 6+7) is a compatible augmentation to the basic MPEG2 frame.

MPEG2 audio encoding process diagram

MPEG-2 audio encoding hardware

To enable monitoring of the encoded audio, an MPEG-2 audio output is available. This output can be used to verify that the encoder settings have been optimised, i.e. the audio is encoded with the required quality and a suitable file size is obtained.

In addition, each channel has an output available for direct feed-through purposes, and for comparison with the encoded audio from the MPEG-2 output.

Dolby, Dolby Surround, Dolby Digital and the Double-D Logo are trademarks of Dolby Laboratories Licensing Corporation

A primer on MPEG-2 compression & DVD Basics with thanks to Philips Industries
What Is DVD??	MPEG-1 / MPEG-2 VIDEO ENCODING	MPEG AUDIO ENCODING