# Improving Perceptual Audio Aesthetic Assessment via Triplet Loss and Self-Supervised Embeddings

arXiv: https://arxiv.org/abs/2509.03292 · HTML: https://mlsanity.com/paper/2509.03292/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 21 of its 22 references that are in the map.
- First posted on arXiv 3 Sep 2025.

## Builds on (most cited first, 21 of 21)

- [FaceNet: A Unified Embedding for Face Recognition and Clustering](https://mlsanity.com/paper/1503.03832/) (2015)
- [Unsupervised Feature Learning via Non-Parametric Instance-level Discrimination](https://mlsanity.com/paper/1805.01978/) (2018)
- [UTMOS: UTokyo-SaruLab System for VoiceMOS Challenge 2022](https://mlsanity.com/paper/2204.02152/) (2022)
- [BEATs: Audio Pre-Training with Acoustic Tokenizers](https://mlsanity.com/paper/2212.09058/) (2022)
- [Meta Audiobox Aesthetics: Unified Automatic Quality Assessment for Speech, Music, and Sound](https://mlsanity.com/paper/2502.05139/) (2025)
- [Generalization Ability of MOS Prediction Networks](https://mlsanity.com/paper/2110.02635/) (2021)
- [The VoiceMOS Challenge 2022](https://mlsanity.com/paper/2203.11389/) (2022)
- [MBNet: MOS Prediction for Synthesized Speech with Mean-Bias Network](https://mlsanity.com/paper/2103.00110/) (2021)
- [Deep Learning-based Non-Intrusive Multi-Objective Speech Assessment Model with Cross-Domain Features](https://mlsanity.com/paper/2111.02363/) (2021)
- [LDNet: Unified Listener Dependent Modeling in MOS Prediction for Synthetic Speech](https://mlsanity.com/paper/2110.09103/) (2021)
- [SpeechLMScore: Evaluating speech generation using speech language model](https://mlsanity.com/paper/2212.04559/) (2022)
- [The VoiceMOS Challenge 2024: Beyond Speech Quality Prediction](https://mlsanity.com/paper/2409.07001/) (2024)
- [Enabling Auditory Large Language Models for Automatic Speech Quality Evaluation](https://mlsanity.com/paper/2409.16644/) (2024)
- [A Study on Incorporating Whisper for Robust Speech Assessment](https://mlsanity.com/paper/2309.12766/) (2023)
- [Fusion of Self-supervised Learned Models for MOS Prediction](https://mlsanity.com/paper/2204.04855/) (2022)
- [Non-Intrusive Speech Intelligibility Prediction for Hearing-Impaired Users using Intermediate ASR Features and Human Memory Models](https://mlsanity.com/paper/2401.13611/) (2024)
- [The VoiceMOS Challenge 2023: Zero-shot Subjective Speech Quality Prediction for Multiple Domains](https://mlsanity.com/paper/2310.02640/) (2023)
- [Self-Supervised Speech Quality Estimation and Enhancement Using Only Clean Speech](https://mlsanity.com/paper/2402.16321/) (2024)
- [Speech foundation models on intelligibility prediction for hearing-impaired listeners](https://mlsanity.com/paper/2401.14289/) (2024)
- [HAAQI-Net: A Non-intrusive Neural Music Audio Quality Assessment Model for Hearing Aids](https://mlsanity.com/paper/2401.01145/) (2024)
- [CORN: Co-Trained Full- And No-Reference Speech Quality Assessment](https://mlsanity.com/paper/2310.09388/) (2023)

## Built on by (most cited first, 1 of 1)

- [Robust Generative Audio Quality Assessment: Disentangling Quality from Spurious Correlations](https://mlsanity.com/paper/2603.16201/) (2026)

## Abstract (arXiv)

We present a system for automatic multi-axis perceptual quality prediction of generative audio, developed for Track 2 of the AudioMOS Challenge 2025. The task is to predict four Audio Aesthetic Scores--Production Quality, Production Complexity, Content Enjoyment, and Content Usefulness--for audio generated by text-to-speech (TTS), text-to-audio (TTA), and text-to-music (TTM) systems. A main challenge is the domain shift between natural training data and synthetic evaluation data. To address this, we combine BEATs, a pretrained transformer-based audio representation model, with a multi-branch long short-term memory (LSTM) predictor and use a triplet loss with buffer-based sampling to structure the embedding space by perceptual similarity. Our results show that this improves embedding discriminability and generalization, enabling domain-robust audio quality assessment without synthetic training data.