With the FCC poised to begin enforcing the CALM Act starting on December 15, some broadcasters are adopting aggressive loudness processing as an easy way to avoid potential fines. According to some industry observers, that approach is unlikely to improve audio quality.
“There are forces at work that encourage using the easiest, least expensive, set-and-forget solutions,” comments Lon Neumann, consulting engineer with Neumann Technologies and a specialist in audio for DTV. “Such solutions may have the unintended consequence of returning us to the dark days of NTSC audio.”
“There’s still this danger of people using too aggressive processing,” agrees Roger Charlesworth, an independent media technology consultant and executive director, DTV Audio Group. “I think the danger is that everybody is coming out of the woodwork who wants to sell something. I think it’s seductive to think you can buy a box and not have to worry about it.”
At FOX Television’s KTTV 11/UPN 13 in Los Angeles, Craig Else primarily mixes news and entertainment shows. Although he mixes to a target loudness level, it can be difficult to anticipate sudden loud noises on live programming, he says.
To keep levels under control, he says, “I have a compressor on all of those channels at about a 3:1 ratio, really fast. If it goes over that, it gets caught either at master control or at the tower. We have brick-wall limiters at those positions.”
That downstream limiting can certainly degrade the audio, he admits: “If you’re mixing a little too hard and you go home and hear the repeat of the show, it’s pumping and doing all sorts of weird things.”
Bruce Arledge, production mixer on Dancing with the Stars and other entertainment shows, strictly adheres to the network loudness specification when mixing, using three sets of meters—a combination of a Linear Acoustics LQ-1000, VU and Durrough peak meters—as a guide. But while Arledge is careful not to exceed ABC’s specified –24 LKFS maximum, commercials sometimes air hotter, he notes.
His hands are tied: “I can’t adjust my mix to that level, because then I’m not adhering to the network standards. There’s nothing worse than getting those memos on Tuesday after the show: ‘Bruce, you hit a –19.’”
The major networks and MVPDs (multichannel video programming distributors) have largely implemented loudness management practices, says Charlesworth. “If we look at NBC or Fox or PBS, they’re all in compliance, they’re processing commercials. Those practices are defined; they’ve figured out how to do them cheaply and scale them efficiently. The heavy lifting in policing it has been done by the MVPDs, Comcast and Time Warner particularly.”
The potential for degraded audio due to aggressive processing is limited, he says: “What remains a concern are mom-and-pop cable companies and smaller stations.”
At the root of the problem, according to Neumann, are the two distinctly different approaches to loudness management taken in North America and Europe and enshrined in their respective recommended practices. In North America, the ATSC standard is based on “the explicit precept that the metadata parameter known as dialnorm must tell the truth about the level of the dialogue,” Neumann says.
In Europe, there may be no associated metadata, as audio is not necessarily encoded into AC-3. Thus, loudness measurements are not centered on dialogue but rather on an average of the entire soundtrack.
The ATSC’s A/85 Recommended Practice further confuses matters in North America with its Annexes J (which addresses short form content) and K (which addresses non-AC-3 codecs). “For long-form [programming], the task is still to measure the loudness of dialogue as the Anchor Element. For short-form, the task is to average the measurement of all content for the duration,” explains Neumann.
In Europe, the EBU R128 document recommends a system of levelgating, which excludes portions of silent and low-level content from the loudness measurement. Dolby Labs, meanwhile, has developed an automated workflow based on its Dialogue Intelligence algorithm.
“This automated system applies seven different tests to audio content and reliably finds the portions of content that are normal dialogue. The result thus derived is then used to control the loudness measurement by dialogue-gating, rather than levelgating as per the EBU [and specified in ITU-R BS.1770-2],” explains Neumann.
A hybrid approach might make the most sense, he says. “There are currently ongoing discussions that suggest that perhaps the best approach would be to use dialogue-gating for long-form measurements, in turn using level-gating for the shortform measurements.”
As a proprietary technology, Dolby’s Dialogue Intelligence was not included as a requirement in the ATSC standard, but it has now been made freely available, reports Neumann. “Now developers everywhere can implement Dialogue Intelligence without royalties due to Dolby. I, for one, strongly encourage all to do so.”