Addressing lingering loudnes challenges

Andrew Sachs

Author: Andrew Sachs

Published 1st December 2013

by Andrew Sachs
Issue 83 - November 2013 Why and how has loudness become such prominent issue?
The loudness of the television commercials aired between program segments has been the source of consumer complaints and subsequent regulatory action in countries around the world. These complaints stem from the networks and advertisers economic incentives for consumers to notice loud advertisers, paired with the expanded dynamic range available with digital audio delivery. Though audio levels remain within a technically acceptable range, advertising may be delivered at higher levels so that consumers notice the louder ads.
Because the factors contributing to perceived loudness are more complex than previous peak metering, the industry took action. The ITU created study groups that attempted to objectify the subjective perception of loudness. These study groups helped to develop ITU-R BS.1770 (algorithms for measuring audio program loudness and true-peak audio level), which dealt with spatial, frequency, and power based components to create an objective measurement of loudness.
The EBU PLOUD groups added to the ITUs efforts by incorporating temporal components (level-based gating) that prevented periods of silence from lowering the loudness measurement. This level-based gating technique would cause a dialog track to measure a consistent loudness regardless of whether there were pauses (silence) between speech elements in the track. The ATSC stayed on a different path by using the dialog (anchor element) power as the indication of the loudness of program in which speech is present. Not coincidentally, this technique is also resistant to the presence of pauses between speech elements.
Additionally, many organizations including EBU and ATSC addressed the difference in loudness perception that can be created by improperly mixing surround (5.1) content so that it is experienced relatively more loudly (up to 3dB) when consumed as a stereo downmix. As 85 percent of HD surround content is consumed on stereo speakers, this stereo downmix measurement is the critical measurement for 5.1-delivered content. This improper mixing technique could be easily abused by advertisers to deliver ads that are compliance in the 5.1 measurement but are up to 3dB when consumed as a downmix.
Practical downstream measurement techniques and tools (e.g., at the output of the set-top box) were not widely utilized. Complicating elements such as content boundary alignments, compression artifacts, and non-automated measurement techniques made it very difficult for broadcasters or networks to look at the end of the broadcast pipe and verify compliance. Without proper measurement techniques and practical widely accepted tolerances, there is bound to be organizational friction dealing with this ambiguity.
While loudness is very easy to legislate (e.g., commercials shall not be louder than the surrounding program), the incentives of broadcasters remain aligned with their advertisers. This, paired with the evolving loudness measurement basis and underutilized downstream measurement techniques, has caused loudness to become a very prominent issue. It is easy to spot but hard to verify. How are broadcasters and other service providers successfully incorporating loudness measurement into their workflows?
Loudness measurement is typically incorporated into broadcasters workflows through several different mechanisms. File-based workflows have file quality assurance (QA) tools that measure asset loudness as one of the many QA elements they perform. If levels are off-target, content is rejected or audio is typically remixed at a different level to match the station target level.
While live event loudness is more challenging because of the dynamics of the event, these events typically benefit from the presence of sound engineers who will mix and level audio according to their ears and to real-time loudness meters. Many recommended practices have wider tolerances for live events because of these challenges.
Finally, broadcasters and networks are measuring the compressed downstream delivered by recording the compressed stream and using automation log records to find the asset boundaries. A loudness meter with content recording then references the content recording and boundaries to measure every asset broadcast for the day. This serves as a final quality check and allows broadcasters to catch holes in loudness control processes while easily delivering on compliance requests.
Loudness control devices were initially installed by broadcasters as sort of a hammer to ensure compliance to loudness legislation. However, as complaints mount regarding the detrimental effect these control devices have on the content experience and as the content supply chain evolves with higher levels of loudness compliance, these control devices are being disabled or taken out entirely. How does a monitoring and logging system actually perform loudness measurement?
One of the primary benefits of using a monitoring and logging system equipped with loudness measurement and reporting functions is that the system can capture and store an array of individual loudness measurements, tied frame-accurately to video, for any number of channels. This affords the operator the ability to dial back and evaluate the content in question, whether it aired a day earlier, a week earlier, or a month ago. Rather than re-run and sample broadcasts for loudness levels, the operator enjoys immediate assurance of compliance.
In both compliance and ad verification applications, operators import the daily as-run logs of all ads into the logging system. These boundaries are paired with the logged content to yield loudness of all aired assets. This information simplifies the location of potential problem areas in logged content and makes it easy to respond to a consumer complaint or regulator query. The operator can very quickly locate and export both audio and video.
Should a viewer complaint or regulatory query demand it, the operator can use the Volicon Observers browser-based interface to navigate quickly through logged content to the ad in question and subsequently export both audio and video, with the real-time loudness parameters burned into the video, into an unambiguous A/V affidavit.
By combining the recorded content, the as-run-log from the automation, and loudness capabilities, the logging system with loudness monitoring enables easy and accurate downstream measurements while providing a simple and convincing affidavit of compliance. What loudness monitoring capabilities are critical today?
Effective loudness measurement requires a system capable of measuring AC-3 dialnorm levels, as well as the appropriate loudness specifications and regulations for a given region or country. For most operators, this will demand a system that accommodates one or more of these broadly accepted standards or specifications: ITU BS.1770-3, ATSC A/85 RP 2011 and 2013, EBU R128 (Tech 3341/2), or ARIB TR-B32. This would require being able to perform ungated measurements, gated measurements, and dialog power (ATSC A/85 anchor element), as well as simultaneous 5.1 and 5.1 downmix measurement. To be fully compliant, a loudness meter should also measure true peak and loudness range (LRA) from the R128 recommendation.
It is also critical that the measurements can be applied on logged content so that precise boundaries and measurement are ensured. Because the measurements are logarithmic, even a small error in time alignment (0.5s) of a commercial boundary could create an error of more than 1dB for a short-duration commercial.
Additionally, for real-time troubleshooting or for performing loudness measurements on commercials when the operator does not know the location of the commercial boundaries because he or she lacks access to the as-run log it is critical that multiple loudness measurements can be performed simultaneously in parallel. (Such measurements might include 10s, 15s, 20s, 30s, 60s, and 120s durations continuously, every 100ms, in sliding windows.) These measurements make it easy for cable to verify compliance of ads for processes such as loudness spot checks without the need for manual recreation of as-run logs from the broadcaster. How are service providers addressing loudness in both surround (5.1) and stereo (2.0) downmix?
Because loudness across both 5.1 and the stereo downmix cannot be properly fixed by simply remixing at a different level, this area of compliance is evolving more slowly than others. To be clear, this is not about the 2.0 downmix supplied with the 5.1 track, but rather about the 2.0 downmix that is created by downmixing the 5.1 track in the consumers home.
Some regulators are, prudently, choosing to regulate the stereo downmix rather than regulate the 5.1. In this case, the assets loudness is only measured and leveled based on the stereo downmix. This is a very practical approach to this challenge, and it ensures a simple solution while optimizing the experience for 85 percent of the consumers of 5.1 content. It also makes the solution easy for broadcasters by clarifying the measurement basis for loudness (and subsequent leveling or control mechanisms).
Other regulators (ATSC) have only noted that the 5.1 downmix should also be measured but are not specific regarding compliance. This is leaving service providers in an ambiguous position of choosing 5.1 only, stereo only, or the louder of the two to use to level the content. As there are no penalties for commercials that are too quiet, you may see providers level according to the loudest.
Because mixing techniques can cause a difference of [+3, -6] on the stereo downmix relative to the 5.1 loudness measurement, this issue has brought many mixing techniques into the spotlight. It is not uncommon to see L, R, and C all contain the same content in a 5.1 dialog mix. The result is a downmix that is 2.8dB hotter and, if left unlevelled, subject to abuse by advertisers. However, penalizing certain track mixes is also perceived as unnecessary interference in the creative process and is slowing resolution of this issue. What steps are you taking to assure that your monitoring solution stay in step with global standards and formats?
We continue to push for recommended practices and legislation that ease compliance efforts so that time and energy can be spent on enhancing quality of the A/V stream and not on interorganizational friction. (Regulators could, for example, mandate the availability of sampled historical as-run logs that would facilitate automated compliance verification for 24 hours of content and skip a lot of error-prone manual processes.)
Volicon participates in global loudness standards bodies and, due to our background in compliance logging, frequently consults with regulators all around the world. This gives Volicon early insight into key measurement and compliance issues and practices, ensuring that we can incorporate them into our product offering. Because Volicon is extremely active in the development and testing of emerging standards for loudness and other aspects of broadcasts, we are well-positioned to adapt our Observer line to meet the current and future requirements facing todays service providers.

Related Listings

Related Articles

Related News

Related Videos

© KitPlus (tv-bay limited). All trademarks recognised. Reproduction of this content is strictly prohibited without written consent.