# Improving Predicted MOS Scores, Not Perceived Quality: Multi-Predictor Test-Time Optimization of Enhanced Speech

arXiv: https://arxiv.org/abs/2609.39028 · HTML: https://mlsanity.com/paper/2609.39028/

## Key facts

- No citing papers in the map yet (data through 19 Oct 2026).
- Placed in Speech enhancement and noise suppression, one of 11 directions in Speech and audio language models.
- Builds on 16 of its 26 references that are in the map.
- First posted on arXiv 30 Sep 2026.

## Placed in

- Speech and audio language models / [Speech enhancement and noise suppression](https://mlsanity.com/speech/speech-enhancement-and-noise-suppression/)

## Builds on (most cited first, 16 of 16)

- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022](https://mlsanity.com/paper/2204.02152/) (2022)
- [DNSMOS P.835: A Non-Intrusive Perceptual Objective Speech Quality Metric to Evaluate Noise Suppressors](https://mlsanity.com/paper/2110.01763/) (2021)
- [NISQA: A Deep CNN-Self-Attention Model for Multidimensional Speech Quality Prediction with Crowdsourced Datasets](https://mlsanity.com/paper/2104.09494/) (2021)
- [SpeechBERTScore: Reference-Aware Automatic Evaluation of Speech Generation Leveraging NLP Evaluation Metrics](https://mlsanity.com/paper/2401.16812/) (2024)
- [SCOREQ: Speech Quality Assessment with Contrastive Regression](https://mlsanity.com/paper/2410.06675/) (2024)
- [Distillation and Pruning for Scalable Self-Supervised Representation-Based Speech Quality Assessment](https://mlsanity.com/paper/2502.05356/) (2025)
- [Towards Low-distortion Multi-channel Speech Enhancement: The ESPNet-SE Submission to The L3DAS22 Challenge](https://mlsanity.com/paper/2202.12298/) (2022)
- [Improving Speech Enhancement with Multi-Metric Supervision from Learned Quality Assessment](https://mlsanity.com/paper/2506.12260/) (2025)
- [ICASSP 2026 URGENT Speech Enhancement Challenge](https://mlsanity.com/paper/2601.13531/) (2026)
- [On the robustness of non-intrusive speech quality model by adversarial examples](https://mlsanity.com/paper/2211.06508/) (2022)
- [Attacking UTMOS: Probing the Robustness of a Speech Quality Assessment Model](https://mlsanity.com/paper/2606.31105/) (2026)
- [GAP-URGENet: A Generative-Predictive Fusion Framework for Universal Speech Enhancement](https://mlsanity.com/paper/2604.01832/) (2026)
- [SLT 2026 REAL-TSE Challenge: Real-world Target Speaker Extraction from Conversational Recordings](https://mlsanity.com/paper/2607.15198/) (2026)
- [A Hybrid Discriminative and Generative System for Universal Speech Enhancement](https://mlsanity.com/paper/2601.19113/) (2026)
- [Technical Report for MERL's Real-TSE Challenge Submission](https://mlsanity.com/paper/2607.09043/) (2026)

## Abstract (arXiv)

Non-intrusive MOS predictors are widely used instead of subjective listening tests to evaluate and rank speech enhancement (SE) systems. If they accurately reflect perceived quality, raising their scores should lead to higher-quality speech. We present the first comprehensive analysis of test-time optimization for the SE task, which directly modifies the enhanced signal to raise the average of multiple MOS predictor scores. On seven systems from the URGENT 2026 challenge, we find that 1)~all the optimized predicted scores increase while reference-based metrics remain nearly unchanged, 2)~a non-optimized predicted score does not increase, and 3)~a MUSHRA listening test shows no improvement in perceived quality. These findings reveal a risk that such optimization can distort evaluations, e.g., biasing comparisons of SE systems regardless of their perceived quality. We believe these findings can inform future evaluation practices: they suggest that predictors used for optimization should not be used for evaluation, and that challenges should keep the predictors used for ranking undisclosed.