Machine Learning: Predicting Audio Quality for high SPL Smartphone Recordings

Guelen P, Zhao D, Terra Pizutti Dos Santos P, Drouadene A, Bacle J. Machine Learning: Predicting Audio Quality for high SPL Smartphone Recordings. In: AES Convention 155. Audio Engineering Society; 2023.pp. 130:1-9. Available from: https://aes.org/publications/elibrary-page/?id=22284

@inproceedings{Guelen2023_22284,
  author = {Guelen, Philippe and Zhao, Dan and Terra Pizutti Dos Santos, Pietro and Drouadene, Arthur and Bacle, Justin},
  title = {{Machine Learning: Predicting Audio Quality for high SPL Smartphone Recordings}},
  booktitle = {AES Convention 155},
  pages = {130:1-9},
  url = {https://aes.org/publications/elibrary-page/?id=22284},
  year = {2023},
  month = oct,
  publisher = {Audio Engineering Society}
}

TY  - CPAPER
TI  - Machine Learning: Predicting Audio Quality for high SPL Smartphone Recordings
AU  - Guelen, Philippe
AU  - Zhao, Dan
AU  - Terra Pizutti Dos Santos, Pietro
AU  - Drouadene, Arthur
AU  - Bacle, Justin
T2  - AES Convention 155
SP  - 1
EP  - 9
M1  - Paper 130
PY  - 2023
DA  - 2023/10/06
UR  - https://aes.org/publications/elibrary-page/?id=22284
PB  - Audio Engineering Society
LA  - en
AB  - In this paper, we explore a machine learning approach to evaluate audio quality for high sound pressure level (SPL) smartphone recordings. Our study is based on perceptual evaluations conducted by technical experts on eight audio sub-attributes (tonal balance, treble, midrange, bass, dynamics, temporal artifacts, spectral artifacts, and other artifacts) of audio quality for 121 smartphones released from 2019 to 2021. To address this task, we propose a Convolutional Neural Network (CNN) model, which proves to be a simple yet effective choice. We employ a pre-augmentation technique to enhance the training dataset size, creating a comprehensive dataset comprising recording spectrograms and corresponding perceptual evaluation scores. Our findings indicate that while the CNN model has certain limitations, it demonstrates promising capabilities in predicting evaluation scores, particularly in aspects of tonal balance, bass, and spectral artifact assessment.
ER  -