An Evaluation Method for Temporal Spatial Sound Attributes

Moiragias G, Mourjopoulos J. An Evaluation Method for Temporal Spatial Sound Attributes. In: AES Convention 154. Audio Engineering Society; 2023. Paper 10644. Available from: https://aes.org/publications/elibrary-page/?id=22051

@inproceedings{Moiragias2023_22051,
  author = {Moiragias, George and Mourjopoulos, John},
  title = {{An Evaluation Method for Temporal Spatial Sound Attributes}},
  booktitle = {AES Convention 154},
  note = {Paper 10644},
  year = {2023},
  month = may,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=22051}
}

TY  - CPAPER
TI  - An Evaluation Method for Temporal Spatial Sound Attributes
AU  - Moiragias, George
AU  - Mourjopoulos, John
T2  - AES Convention 154
M1  - Paper 10644
PY  - 2023
DA  - 2023/05/06
UR  - https://aes.org/publications/elibrary-page/?id=22051
PB  - Audio Engineering Society
LA  - en
AB  - This work proposes a methodology for assessing temporal sound attributes, which can be applied to register time-varying perceptual attributes of audio signals reproduced via di?erent spatial systems. The proposed methodology is inspired by state of the art temporal sensory evaluation of food, and such concepts are integrated into current spatial sound evaluation approaches. Here, as an example, we consider the dynamic evaluation of the perceptual attribute of envelopment, time intensity of envelopment (TIE), for music samples reproduced in three di?erent spatial formats (mono, stereo and surround 5.0). The work describes a proposal for appropriate listening tests and data collection, along with a novel method for post-screening and extracting reliable temporal attributes from the listener assessments. A thorough description of each component of the listening test is given and a novel methodology is proposed for assessing the agreement of the listening panel and reducing the noise of the obtained data, based on non-centered Principal Component Analysis and hierarchical clustering algorithms.
ER  -