A Model of Distraction in an Audio-on-Audio Interference Situation with Music Program Material

Francombe J, Mason R, Dewhirst M, Bech S. A Model of Distraction in an Audio-on-Audio Interference Situation with Music Program Material. J Audio Eng Soc. 2015;63(1/2):63-77. doi:10.17743/jaes.2015.0006

@article{Francombe2015_17567,
  author = {Francombe, Jon and Mason, Russell and Dewhirst, Martin and Bech, Søren},
  title = {{A Model of Distraction in an Audio-on-Audio Interference Situation with Music Program Material}},
  journal = {Journal of the Audio Engineering Society},
  volume = {63},
  number = {1/2},
  pages = {63--77},
  year = {2015},
  month = jan,
  publisher = {Audio Engineering Society},
  doi = {10.17743/jaes.2015.0006},
  url = {https://doi.org/10.17743/jaes.2015.0006}
}

TY  - JOUR
TI  - A Model of Distraction in an Audio-on-Audio Interference Situation with Music Program Material
AU  - Francombe, Jon
AU  - Mason, Russell
AU  - Dewhirst, Martin
AU  - Bech, Søren
T2  - Journal of the Audio Engineering Society
J2  - J. Audio Eng. Soc.
VL  - 63
IS  - 1/2
SP  - 63
EP  - 77
PY  - 2015
DA  - 2015/01/06
DO  - 10.17743/jaes.2015.0006
UR  - https://doi.org/10.17743/jaes.2015.0006
PB  - Audio Engineering Society
LA  - en
AB  - There are many situations in which multiple audio programs are replayed over loudspeakers in the same acoustic environment, allowing listeners to focus on their desired target program. Where this situation is deliberately created and the different program items are centrally controlled, each listener can be viewed as having a personal sound zone system. In order to evaluate and optimize such situations in a perceptually relevant manner, the authors created a predictive model using the features that contribute to the distraction from unwanted sounds. Feature extraction was motivated by a qualitative analysis of subject responses. Distraction ratings were collected for one hundred randomly created audio-on-audio interference situations with music target and interferer programs. The selected features were related to the overall loudness, loudness ratio, perceptual evaluation of audio source separation, and frequency content of the interferer. The model was found to predict accurately for the training and validation datasets.
ER  -