The calm before the storm.. the inside of the Staples Center during rehearsals.

Transmission and Delivery

At this point, there are now two complete soundtracks – the final high-end discrete multichannel mix produced by Paul Sandweiss in the OSR trailer, and Ed Greene’s Lt/Rt matrix mix for the rest of the world that can’t receive 5.1 digital audio. Both mixes, along with the high-definition video signal, are fed to another on-site truck housing the CBS transmission system. To get to the CBS broadcast center in New York, the audio mixes are encoded into Dolby E.

Unlike the Dolby Digital delivery format, also known as AC-3, used in consumer audio applications, Dolby E is a more robust encode/decode (CODEC) process designed for use in professional broadcasting, where the source material travels through multiple paths of decoding and re-encoding, and undergoes other transformations that only occur in distribution environments. By design, Dolby E can be encoded and decoded many times without loss of quality, and uses much lighter compression than AC-3 – generally about four-to-one compression, although the exact data rates per channel may vary.

Dolby E can carry 8 discrete channels of information within one AES pair. In typical use, that comprises a 5.1 signal and a two-channel signal (usually a Pro Logic II mix, but it could also be used for a second language mix or other applications).

In the CBS trailer, which is equipped with an MPEG video encoder, Dolby E encoder, and transmission gear, the discrete 5.1 and two-channel surround mixes are encoded into a Dolby E stream, which is then multiplexed with the high def video signal into the complete MPEG transport stream that gets transmitted via satellite link to the CBS broadcast center in New York. (No standard definition video feed is produced; it is downsampled from the high definition signal.)

The Dolby E stream also contains critical metadata, separate from the actual audio data, which contains information about the audio data, and how to correctly perform encoding and decoding at each stage. The metadata doesn’t alter audio content – parts of it are used in different places along the way: the Dolby E ‘program configuration’ (5.1 + 2), some Dolby Digital encoding parameters, and the decoding process in the consumer’s living room. All metadata is authored onsite at the Grammy show by the CBS engineer doing the Dolby E encoding, and is carried with the Dolby E stream to the New York broadcast center. To complicate the picture, CBS has invested in developing technology to redundantly embed the metadata in the video stream’s vertical ancillary space (otherwise known as VANC), along with other data used for video transmission. The metadata, therefore, ends up being carried in both the VANC and Dolby E streams. CBS recommends that stations access the metadata through VANC, but those without VANC decoders can extract it from the Dolby E decoder output.

	Just one of 1,000 micophone feeds at the Grammy Awards.

In New York, the Dolby E stream is decoded into the separate audio mixes and accompanying metadata. At this point, the network can integrate the feeds with the rest of its programming – inserting commercials, etc., as well as vetting the content (the broadcast is delayed five seconds as a precaution against wardrobe and vocabulary malfunctions). When all network steps have been completed, the robust nature of Dolby E allows the audio, video, and metadata to be re-encoded without loss of quality and sent out via satellite to local stations and other carriers such as DirecTV and Dish Network.

Stations capable of handling 5.1 audio receive the MPEG/Dolby E stream and extract the audio, video, and metadata. Through their local processing infrastructure, stations can insert commercials and other content, switching back and forth between the network and local programming. At the final stage, the stations re-encode the 5.1 audio into an AC-3 signal using the Dolby Digital DP569 encoder. Some of the metadata is used by the DP569 to govern the AC-3 encoding, the rest is passed through with the AC-3 stream to the home. The consumer’s Dolby Digital-capable home audio receiver or processor uses the metadata to correctly apply dialnorm, number of channels, dynamic range compression, and other parameters designed to help the program sound as good as possible on a wide range of playback systems – from high end home theaters to the speakers built into a television. All of these instructions are authored back in the CBS transmission trailer, and stay with the signal from the content creation site, over the satellite to the broadcast center, over satellite again to the local high-def station or carrier, and out to the homes of consumers with 5.1 audio capability. For the majority of standard-def viewers watching the Grammys via the traditional analog television path, there’s always Ed Greene’s surround-encoded Lt/Rt mix.

And that, boys and girls, is how a stadium-full of sound gets to be the Grammy surround telecast you hear in your living room.

Special thanks to Rocky Graham of Dolby Labs for talking me through the signal transmission cycle, to Robbie Clyne of Neilson/Clyne Public Relations for arranging HFR’s backstage tour and Ingrid Powell of the Recording Academy for her invaluable help.


Philip Brandes (Text) and Steve Grayson (Images) - 21/02/2005



The text contents of this feature are exclusive to and the sole property of High Fidelity Review, copyright 2005, all rights reserved. This feature cannot be reproduced in whole or in part without the written permission of High Fidelity Review. Can’t see the images on this page? Click here.