Perceptual Evaluation of Machine Learning and Non-ML Emulations of the Vox AC30 Amplifier

Vallejo M, McLoughlin M, Kearney G. Perceptual Evaluation of Machine Learning and Non-ML Emulations of the Vox AC30 Amplifier. In: 2025 AES International Conference on Artificial Intelligence and Machine Learning for Audio. Audio Engineering Society; 2025. Paper 30. Available from: https://aes.org/publications/elibrary-page/?id=23019

@inproceedings{Vallejo2025_23019,
  author = {Vallejo, Mario and McLoughlin, Michael and Kearney, Gavin},
  title = {{Perceptual Evaluation of Machine Learning and Non-ML Emulations of the Vox AC30 Amplifier}},
  booktitle = {2025 AES International Conference on Artificial Intelligence and Machine Learning for Audio},
  note = {Paper 30},
  year = {2025},
  month = sep,
  publisher = {Audio Engineering Society},
  url = {https://aes.org/publications/elibrary-page/?id=23019}
}

TY  - CPAPER
TI  - Perceptual Evaluation of Machine Learning and Non-ML Emulations of the Vox AC30 Amplifier
AU  - Vallejo, Mario
AU  - McLoughlin, Michael
AU  - Kearney, Gavin
T2  - 2025 AES International Conference on Artificial Intelligence and Machine Learning for Audio
M1  - Paper 30
PY  - 2025
DA  - 2025/09/02
UR  - https://aes.org/publications/elibrary-page/?id=23019
PB  - Audio Engineering Society
LA  - en
AB  - Twenty-two listeners evaluated the perceptual similarity of three commercially available emulations of a Vox AC30 amplifier to the original hardware reference under controlled conditions: (1) a machine-learning plugin trained on the exact test amplifier, (2-3) two non-ML plugins modeling unspecified AC30 units. Using MUSHRA methodology, results showed the unit-specific ML plugin achieved perceptual indistinguishability from the hardware reference in 1/3 cases and outperformed both non-ML alternatives in 5/6 pairwise comparisons. This demonstrates that while non-ML plugins necessarily generalize across hardware units, ML techniques can achieve high perceptual fidelity when trained on target amplifiers. This represents a previously unattainable capability with significant implications for audio preservation and production.
ER  -