Where's the remote? The on-going development of loudness monitoring and measurement


Television sound is often a more common cause of complaint from viewers than picture quality. Audibility - or intelligibility - has been a particular issue in recent years, with mumbling actors or dialogue drowned out by music and effects. A longer-standing problem is loudness, which forces people to keep adjusting the volume on their sets because it fluctuates so much.

Loudness is the discrepancies in sound levels between one audio signal and another. The most recognisable instance is the transition from a drama programme to a commercial or interstitial. In the days when metering was on either PPMs (peak programme meters) or VU (volume unit) meters, the main show could register the same peak level at a randomly selected point as the ad or promo. But because dramas have a much wider dynamic range and tend not to be as heavily compressed as commercials, they can sound quieter.

This is known as 'perceived volume', meaning we hear sounds at different levels despite what an audio meter tells us. Loudness can also occur within a programme when the dynamic range goes from, for example, gentle country sounds to loud explosions and a full orchestra score. Either way it meant viewers had to keep adjusting the volume, which, in the days before remote controls, forced them to get up out of the armchair and walk across the room to the TV.



While loudness was the bane of viewers' lives for many years, broadcast engineers were striving to find ways to deal with it. During the 1950s the BBC's R&D department identified loudness as a key area of research. UK commercial broadcaster Thames TV's own R&D team began looking into it in the 1970s, resulting in an algorithm that was put to practical use in one of the first loudness meters in the 1980s.

Competing products appeared during the 1990s but the proliferation in TV services through the growth in satellite and digital broadcasting compounded the problem, leading standards bodies to take action. In 2003 the ITU (International Telecommunications Union) set up a working group that in 2006 and 2007 produced ITU-R BS1770 (algorithms to measure audio programme loudness and true-peak audio level) and ITU-R BS1771 (requirements for loudness and true-peak indicating meters). These formed the basis of new guidelines for broadcasters and post-production facilities and were a good start.

However, they only specified the algorithm and not other important parameters such as the target level and where the common loudness level should be set. Other broadcast and standards bodies decided to build on the ITU's work and produce recommendations with more specific targets and operational practices.

The European Broadcasting Union (EBU), through the PLOUD working group, produced R128 in 2010. The core of this specified loudness normalisation and the permitted maximum level of audio signals, plus a method for gating. In the US the ATSC (Advanced Television Systems Committee) published A/85, which led to legislation in the form of the 2010 CALM (Commercial Advertisement Loudness Mitigation) Act. Similar standards followed in Japan (ARIB TR-B32) and Australia (OP-59).



Meter manufacturers and designers of specialist software adopted some or all of the new standards. This enabled dubbing mixers to create compliant soundtracks through mixing on new hardware metering units in conjunction with their ears and experience; at the same time file-based media could be checked throughout the distribution process using dedicated programs.

The result was a noticeable difference on-air and a dramatic decrease in the number of complaints to broadcasters and regulators. It should not be underestimated how much progress has been made in monitoring and managing loudness in the last 14 years. There was most definitely a problem in the years prior, which is why the ITU and PLOUD working groups were created. The results of their work - BS1770 and R128, as well as A/85 and the other standards - produced tools that went a long way towards solving the problem. But we are not there yet.

Since the introduction of R128, for example, the broadcast landscape has change considerably. HD has become part of the channel offering, while streaming is now a major way for people to watch films and TV programmes. Both of these have brought new loudness problems, which have led to some new standards - for example AES TD1004.1.15-10, recommendations for loudness of audio streaming and network file playback - and proposed revisions to existing ones. R128 has been revised since its launch and is now on its fourth iteration; a new version addressing streaming is due this year.

There has also been a significant change in the way people watch what we now loosely call television. There is more viewing on mobile phones but within the home there has also been a considerable shift. When R128 came out the old fashioned TV in a box, based on the cathode ray tube (CRT), was still a fixture in many living rooms. This had two integral loudspeakers, which produced reasonably good audio quality. Home cinema systems, delivering 5.1 surround, have always been the preserve of enthusiasts or the rich but do deliver a higher audio experience.

Today the CRT is dead as a domestic device and the flat screen is the ubiquitous display. While these have increasingly delivered better images, their audio performance is fairly poor. Consequently, even average viewers now hook up soundbars to their flat screens for a better sound. Because of this it seems that what progress was made has been counter-balanced by people becoming more sensitive to loudness control.

Viewers are particularly aware of the difference in levels between different channels, a discrepancy that is apparent when switching from digital terrestrial services to streaming. As a result manufacturers, like TSL Products, have continued development of loudness monitoring equipment, not just to keep in line with the standards, although those have provided a valuable foundation for R&D, but also in response to changes in distribution techniques and viewing habits.

Loudness has always been a hot topic at TSL Products. We have produced a series of measuring devices over the years for the critical playout and transmission sectors. Our latest, which was scheduled to receive its big launch at the NAB Show, is the SAM-Q-SDI. TSL Products considers itself a pioneer in loudness monitoring and SAM-Q-SDI is the result of our continuing research into this specialised and crucial field.

Externally the SAM-Q-SDI is very similar to previous TSL loudness units. Internally the big difference is the inclusion of eight loudness probes that are able to measure eight separate programmes simultaneously. As would be expected of a modern loudness monitor, it supports short-term, momentary and integrated values. This information can be read on the front panel display like a conventional meter but data can also be logged and monitored across Ethernet networks to digital displays.

SAM-Q-SDI is a 1RU 19-inch rack-mounted unit aimed at playout centres and master control rooms. It is able to handle a wide range of audio formats, from mono right up to 9.1+4. The monitor works to R128 Part IV and is 80 percent software and 20 percent hardware. This makes it very different to TSL monitors of the past and means it can be easily updated as and when new versions of R128 are released.

Loudness is a problem that will always exist in TV broadcast but through research, the latest technology and evolving standards, we are definitely better at dealing with it.