Evaluating Live Loudness Meters from Engineers’ Actions and Resulting Output Levels

Allan J, Berg J. Evaluating Live Loudness Meters from Engineers’ Actions and Resulting Output Levels. J Audio Eng Soc. 2018;66(7/8):556-577. doi:10.17743/jaes.2018.0029

@article{Allan2018_19706,
  author = {Allan, Jon and Berg, Jan},
  title = {{Evaluating Live Loudness Meters from Engineers’ Actions and Resulting Output Levels}},
  journal = {Journal of the Audio Engineering Society},
  volume = {66},
  number = {7/8},
  pages = {556--577},
  year = {2018},
  month = jul,
  publisher = {Audio Engineering Society},
  doi = {10.17743/jaes.2018.0029},
  url = {https://doi.org/10.17743/jaes.2018.0029}
}

TY  - JOUR
TI  - Evaluating Live Loudness Meters from Engineers’ Actions and Resulting Output Levels
AU  - Allan, Jon
AU  - Berg, Jan
T2  - Journal of the Audio Engineering Society
J2  - J. Audio Eng. Soc.
VL  - 66
IS  - 7/8
SP  - 556
EP  - 577
PY  - 2018
DA  - 2018/07/06
DO  - 10.17743/jaes.2018.0029
UR  - https://doi.org/10.17743/jaes.2018.0029
PB  - Audio Engineering Society
LA  - en
AB  - Loudness discrepancies in television and radio frequently produce listener annoyance. Variations in loudness can be traced to the use of quasi-PPM based audio level meters in conjunction with different amounts of compression dynamics. A satisfactory live loudness meter should assist the engineer to: (a) achieve the recommended target level for a program, (b) compensate for content-dependent delimited offsets in loudness, and (c) compensate for fast changes in loudness. This paper investigates how the ballistic properties of live loudness meters affect the engineers’ actions with fader position and the resulting output levels. In order to explore the quality of loudness meters, the researchers simulated a live broadcast show with mixing engineers who had different degrees of experience. The resulting output levels were analyzed and interpreted using a linear mixing model. The results showed that the meters with the slower integration times produced less dispersion of output levels for parts of the program. Varying integration times of the meters did not cause a significant difference in the reaction time.
ER  -