Getting Vocal About Loudness
By Dan Daley
SVG Audio Editor
The first in a series of articles looking at the challenge of loudness levels as it relates to sports programming.
Loud audio during a TV broadcast has been a consumer complaint since the days of the Bert & Harry Piels beer ads on
The Honeymooners that were often louder than Ralph and Alice’s arguments. The problem persists and festers in the digital age, when more bandwidth simply allows even greater level disparities between various programs. Broadcast-sports audio has its own specific problems in this regard.
The industry has come a long way since Bert & Harry. The introduction of Dolby’s LM-100 meter created a way to achieve a long-term integrated measurement. Because the diversity of potential audio elements in a typical sports broadcast can overwhelm the announcers, Dolby’s Dialnorm uses dialogue as the target cue from which to precisely optimize the rest of a program’s audio.
Next, the ITU, through the BS.1770 standard, has taken input from broadcast and loudness experts and created an international recommendation that manufacturers can support. This has created a baseline upon which companies are offering new approaches.
“It’s one thing to manage loudness in a postproduction situation but something else and much more difficult in a live broadcast,” says Dave Moulton, a Boston-area acoustics-measurement and loudspeaker-design specialist, who cites the wide range of supporting noises found in live sporting events — from cow bells to Thunder Sticks — that can foil algorithmic-based attempts to balance sound.
Linear Acoustic’s AERO transmission-loudness manager system includes a feature that addresses just that. Its Crowd Control dialog-protection processing neutralizes, prior to transmission, the effects of home-user–selectable effects, such as the “Hyper Surround” mode that can mask the voiceover as crowd sounds end up too loud in the surround channels.
However, Tim Carroll, founder and president of Linear Acoustic, says loudness is only part of the issue. “Dynamic range is the other difficult part,” he says. “A digital delivery system that reaches all the way to consumers, coupled with its accompanying metadata, has the ability to keep loudness and dynamic range as the two separate issues they actually are. Unfortunately, with [DTV] comes some challenges, as an evening of programming that is loudness-matched could also be too dynamic for many viewers.”
Neural Audio’s Loudness Control (NLC) bases its approach on perceived loudness with processing that mimics the human auditory response by detecting spectral and density differences, inter-channel relationships, and temporal overlaps within the audio signal. After detection, NLC applies the appropriate gain or attenuation to the audio to achieve a user-defined loudness level.
Neural Audio chief scientist James “JJ” Johnston notes that loudness has been able to be measured for some time with equal-loudness formulas, such as the Fletcher-Munson Curve, but that signal spectrum, as well as signal intensity or SPL, has to be vetted and the signal analyzed on at least a “critical-band” resolution level.
“For sports audio, we know how to do that for mono broadcasts and pretty well for stereo,” says Johnston, who has written extensively on the matter. “For 5.1 and 7.1 multichannel broadcasts, today’s mixes are just starting to scratch the surface.”