special report audio processing
correction is an important processing function, which can consist of multi- band and wideband dynamic range adjustment and level shifting. Automatically upmixing from stereo
to 5.1 surround maintains a consistent on-air channel format. And performing a corresponding downmix allows stereo-only devices to use the same surround signal as the main broadcast. Using these processors, which are available from a number of manufacturers, can ensure that all audio passing through a broadcast facility has consistent average loudness and encoding even in our non-ideal world.
Transport stream processing
A transport stream (TS) is intended to transmit single or multiple programmes, multiplexed programmes along with timing references and supporting metadata. These separate programmes can be transmitted at different bit rates, either constant or variable. The total bit rate of the transport stream is a constant, however. This is achieved by inserting null (zero value) packets as needed to pad the transport stream bit rate. Some of the supporting data is
structurally oriented, such as the programme clock reference (PCR) and packet identification (PID) values. These are used respectively to reconstruct a local clock at the decoder, and to uniquely identify all the packets of a similar programme or other metadata (‘table’ data) in the stream. This table data describes which individual programmes are associated together (such as video and its audio), and what PID different values in the stream are. Some of these tables are defined by the MPEG specification and some by particular broadcast standards, such as DVB or ATSC. Carrying audio signals on a multiple
programme transport stream offers many advantages. One of the major ones is its high signal density. A large number of video programmes can be carried, and each video can have multiple audio programmes associated with it. Each video can be transmitted at a different rate, as can each audio, and different audio bit rate reduction techniques can be used for different programmes. Some installations have no choice
but to accept their content as transport streams, as that is how their network chooses to distribute it.
58 l ibe l november/december 2011 l
www.ibeweb.com There is another advantage to using
Processing equipment is now becoming available that will allow audio processing in the transport stream itself, allowing higher processing and rack space densities. And future equipment should increase these processing densities as well as allowing transport over Gigabit Ethernet, further blurring the broadcast signal/ information technology line.
an MPTS related to the packetised nature of the programmes it contains. Once the video and audio data are packetised and not continuous, more choices are available with the physical transport layer. It no longer requires video-specific cabling and connectors - Gigabit Ethernet is perfectly suited to carrying a MPTS. Equipment designed for Gigabit Ethernet is certainly more generic and more widely available than video and audio specific gear. All of the basic audio processing
functions can be performed on the audio programmes in a multiple programme transport stream. Equipment and techniques for finding the audio programmes and extracting them are well known and reliable. The equipment currently required is, however, anything but efficient from a space and power standpoint. And even more critically, some standard audio processing has the ability to alter data in the transport stream other than the audio itself. To process audio with current
generation equipment requires a number of steps both before and after the processor itself. The transport stream must first be fed to a MPEG decoder to extract the desired audio (and video if needed). The extracted audio can then be processed as required, using as many processors as determined by the channel count. The processed audio can then be re- encoded and multiplexed back into the original transport stream. This re- insertion presents some issues, namely that the unprocessed data may change to some degree. If the same number of channels has changed because of an upmix or downmix, descriptive table data will be changed accordingly. If the same number of channels is re-inserted but the data rates have changed, some of the original transport stream information - at a minimum, the null packets - will have to be altered. A minimum of three pieces of equipment are needed to process any number of audio channels - a demultiplexer/decoder, the processor itself, and an encoder/remultiplexer. In the ideal situation this processing
would be performed on the transport stream itself. With a transport stream input and output the need for external encoders, decoders, and multiplexers is eliminated. Single programmes can be identified, extracted, processed and reinserted all within a single device. All the standard audio processing steps could be replicated in
this device, such as upmixing, decoding, and loudness correction. Ideally these processing steps would be performed in the bit rate-reduced space, eliminating the recoding step and potential reduction of audio quality. The loudness processing steps should be ‘non-destructive’, ie maintain the ability to revert (under viewer control at the home) to the full dynamic range of the original content, thus being better suited to a wide range of listening environments. The capacity for higher density also
exists if there is enough computational horsepower within the processing device, allowing multiple PIDs to be processed and reinserted. An ideal device would also be able to accommodate transport streams in Gigabit Ethernet in addition to the currently used ASI format. Equipment that can process transport streams in this manner is just starting to appear. Until then there are steps that can
be taken to process audio carried in a transport stream as efficiently as possible. The first step is to pick a processor that actually sounds good while processing the type of audio that will pass thorough it in daily use - nothing can ever replace a listening test for audio processing gear. Secondly, ensure the processor has physical interfaces to match the decoders and encoders used elsewhere. Setting up the decoders and
encoders is another important aspect. Ideally the processed audio will be re- inserted into the transport stream at the same bit rate as it was extracted. This minimises the impact the remultiplexing process will have on other programmes and metadata contained in the transport stream. Obviously, in a densely populated transport stream, increasing the bit rate before reinsertion could over fill the transport stream capacity. Multiple programme transport
streams are a widely used and well understood means of efficiently carrying video and audio signals. With the proper precautions, standard audio processing can be performed using currently available equipment. Processing equipment is now becoming available that will allow audio processing in the transport stream itself, allowing higher processing and rack space densities. And future equipment should increase these processing densities as well as allowing transport over Gigabit Ethernet, further blurring the broadcast signal/information technology line.
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60