Perceptual Evaluation of the MPEG-I Immersive AudioStandard

Silzle A, Delgado PM, Restrepo Garcia A, Disch S, Terentiv L, Karlsson E, Leppanen J, Hiroyuki E, Jelfs S. Perceptual Evaluation of the MPEG-I Immersive AudioStandard. In: Express Paper. Audio Engineering Society; 2026.pp. 407:1-9. Available from: https://aes.org/publications/elibrary-page/?id=23149

@inproceedings{Silzle2026_23149,
  author = {Silzle, Andreas and Delgado, Pablo M. and Restrepo Garcia, Alejandro and Disch, Sascha and Terentiv, Leon and Karlsson, Erlendur and Leppanen, Jussi and Hiroyuki, Ehara and Jelfs, Sam},
  title = {{Perceptual Evaluation of the MPEG-I Immersive AudioStandard}},
  pages = {407:1-9},
  url = {https://aes.org/publications/elibrary-page/?id=23149},
  year = {2026},
  month = may,
  publisher = {Audio Engineering Society}
}

TY  - CPAPER
TI  - Perceptual Evaluation of the MPEG-I Immersive AudioStandard
AU  - Silzle, Andreas
AU  - Delgado, Pablo M.
AU  - Restrepo Garcia, Alejandro
AU  - Disch, Sascha
AU  - Terentiv, Leon
AU  - Karlsson, Erlendur
AU  - Leppanen, Jussi
AU  - Hiroyuki, Ehara
AU  - Jelfs, Sam
SP  - 1
EP  - 9
M1  - Paper 407
PY  - 2026
DA  - 2026/05/28
UR  - https://aes.org/publications/elibrary-page/?id=23149
PB  - Audio Engineering Society
LA  - en
AB  - The recently finalized ISO international standard (IS) on MPEG-I Immersive Audio enables interactive sixdegrees-of-freedom (6DoF) audio rendering for a multitude of virtual-reality and augmented-reality (VR/AR) acoustic scenarios and applications. It consists of comprehensive modeling of room acoustics and intricate acoustic phenomena, including e.g. occlusion, reflection, transmission and diffraction caused by sound obstacles, Doppler effect, and dynamic environment changes triggered by user interactivity. This paper describes concept, methodology and results of the final verification test of the standard. The perceptual quality of the renderer was assessed, stabilized by perceptual anchors, in an interactive listening test using different in- and outdoor acoustic scenes. The considerations behind the selection of the test setup and the design of suitable anchors will be discussed. In the test, 56 subjects participated distributed across six labs using the ITU-R Rec. BS.2132 multi-stimulus method on a 100-point scale for three conditions (IS, mid- and low-anchor) in 10 VR scenes plus two repetitions. The finally selected mid and low anchors have demonstrated stable quality across diverse scenes with progressive timbre and spatial degradations, and the results show a clear separation of the conditions (IS > mid > low); the low anchor was stable (around 16 points median value) while the mid anchor varied by scene (around 47 points). Absolute ratings are scene dependent and their quartile range for some scenes can exhibit up to 20 points. The median value obtained for the IS varied slightly per lab, indicating that apparently some labs had more critical listeners than others. The IS is rated with a median of 84 points among all labs, which is the excellent region of the scale.
ER  -