The T12 System For Audiomos Challenge 2025: Audio Aesthetics Score Prediction System Using KAN- And Versa-based Models
2025 · Katsuhiko Yamamoto, Koichi Miyazaki, Shogo Seki
Abstract
We propose an audio aesthetics score (AES) prediction system by CyberAgent (AESCA) for AudioMOS Challenge 2025 (AMC25) Track 2. The AESCA comprises a Kolmogorov--Arnold Network (KAN)-based audiobox aesthetics and a predictor from the metric scores using the VERSA toolkit. In the KAN-based predictor, we replaced each multi-layer perceptron layer in the baseline model with a group-rational KAN and trained the model with labeled and pseudo-labeled audio samples. The VERSA-based predictor was designed as a regression model using extreme gradient boosting, incorporating outputs from existing metrics. Both the KAN- and VERSA-based models predicted the AES, including the four evaluation axes. The final AES values were calculated using an ensemble model that combined four KAN-based models and a VERSA-based model. Our proposed T12 system yielded the best correlations among the submitted systems, in three axes at the utterance level, two axes at the system level, and the overall average. We also
Authors
(none)
Tags
Stats
Related papers
- The T05 System For The Voicemos Challenge 2024: Transfer Learning From Deep Image Classifier To Naturalness MOS Prediction Of High-quality Synthetic Speech (2024)0.00
- SAMOS: A Neural MOS Prediction Model Leveraging Semantic Representations And Acoustic Features (2024)2.26
- MMMOS: Multi-domain Multi-axis Audio Quality Assessment (2025)0.00
- Using Rater And System Metadata To Explain Variance In The Voicemos Challenge 2022 Dataset (2022)5.84
- Automos: Learning A Non-intrusive Assessor Of Naturalness-of-speech (2016)0.00
- SZU-AFS Antispoofing System For The Asvspoof 5 Challenge (2024)5.24
- The Kriston AI System For The Voxceleb Speaker Recognition Challenge 2022 (2022)0.00
- DDOS: A MOS Prediction Framework Utilizing Domain Adaptive Pre-training And Distribution Of Opinion Scores (2022)9.03