Assessing the relevance of perceptually driven objective metrics in the presence of handling noise

Angonin C, Theofanis Chourdakis E, Åeng RA. Assessing the relevance of perceptually driven objective metrics in the presence of handling noise. In: AES Convention 152. Audio Engineering Society; 2022. Paper 10580. Available from: https://aes.org/publications/elibrary-page/?id=21693

@inproceedings{Angonin2022_21693,
  author = {Angonin, Céline and Theofanis Chourdakis, Emmanouil and Åeng, Ruben Andre},
  title = {{Assessing the relevance of perceptually driven objective metrics in the presence of handling noise}},
  booktitle = {AES Convention 152},
  note = {Paper 10580},
  year = {2022},
  month = may,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=21693}
}

TY  - CPAPER
TI  - Assessing the relevance of perceptually driven objective metrics in the presence of handling noise
AU  - Angonin, Céline
AU  - Theofanis Chourdakis, Emmanouil
AU  - Åeng, Ruben Andre
T2  - AES Convention 152
M1  - Paper 10580
PY  - 2022
DA  - 2022/05/06
UR  - https://aes.org/publications/elibrary-page/?id=21693
PB  - Audio Engineering Society
LA  - en
AB  - This paper examines how perceptually driven objective metrics found in the speech enhancement and separation literature react when adding handling noise to speech corrupted with environmental noise. Identifying sensitive metrics will inform us which metrics are appropriate for the development or evaluation of speech enhancement techniques when dealing with handling noise. Using an in-house synthetic dataset and paired sample tests, we examine how nine different perceptual metrics behave on audio mixtures containing both handling and background noise. We show that eight of them react to handling noise but only when the handling to background noise power ratio is over a specific threshold which we identify using logistic regression.
ER  -