ASTAR-NTU solution to AudioMOS Challenge 2025 Track1

TL;DR

This paper presents a winning system for the AudioMOS 2025 Challenge that automatically predicts music impression and text alignment using a dual-branch architecture with pre-trained models, achieving high correlation scores.

Contribution

The paper introduces a novel dual-branch architecture with cross-attention and soft label encoding for improved automatic evaluation of music quality and alignment.

Findings

Achieved SRCC of 0.991 for music impression prediction.

Achieved SRCC of 0.952 for text alignment prediction.

Significant improvement over baseline scores.

Abstract

Evaluation of text-to-music systems is constrained by the cost and availability of collecting experts for assessment. AudioMOS 2025 Challenge track 1 is created to automatically predict music impression (MI) as well as text alignment (TA) between the prompt and the generated musical piece. This paper reports our winning system, which uses a dual-branch architecture with pre-trained MuQ and RoBERTa models as audio and text encoders. A cross-attention mechanism fuses the audio and text representations. For training, we reframe the MI and TA prediction as a classification task. To incorporate the ordinal nature of MOS scores, one-hot labels are converted to a soft distribution using a Gaussian kernel. On the official test set, a single model trained with this method achieves a system-level Spearman's Rank Correlation Coefficient (SRCC) of 0.991 for MI and 0.952 for TA, corresponding to a relative improvement of 21.21\% in MI SRCC and 31.47\% in TA SRCC over the challenge baseline.