# Fusion of Self-supervised Learned Models for MOS Prediction

arXiv: https://arxiv.org/abs/2204.04855 · HTML: https://mlsanity.com/paper/2204.04855/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 29 references that are in the map.
- First posted on arXiv 11 Apr 2022.

## Builds on (most cited first, 13 of 13)

- [wav2vec 2.0: A Framework for Self-Supervised Learning of Speech Representations](https://mlsanity.com/paper/2006.11477/) (2020)
- [HuBERT: Self-Supervised Speech Representation Learning by Masked Prediction of Hidden Units](https://mlsanity.com/paper/2106.07447/) (2021)
- [WavLM: Large-Scale Self-Supervised Pre-Training for Full Stack Speech Processing](https://mlsanity.com/paper/2110.13900/) (2021)
- [data2vec: A General Framework for Self-supervised Learning in Speech, Vision and Language](https://mlsanity.com/paper/2202.03555/) (2022)
- [ESPnet: End-to-End Speech Processing Toolkit](https://mlsanity.com/paper/1804.00015/) (2018)
- [MOSNet: Deep Learning based Objective Assessment for Voice Conversion](https://mlsanity.com/paper/1904.08352/) (2019)
- [Voice Conversion Challenge 2020: Intra-lingual semi-parallel and cross-lingual voice conversion](https://mlsanity.com/paper/2008.12527/) (2020)
- [Generalization Ability of MOS Prediction Networks](https://mlsanity.com/paper/2110.02635/) (2021)
- [The Voice Conversion Challenge 2018: Promoting Development of Parallel and Nonparallel Methods](https://mlsanity.com/paper/1804.04262/) (2018)
- [The VoiceMOS Challenge 2022](https://mlsanity.com/paper/2203.11389/) (2022)
- [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)
- [Neural MOS Prediction for Synthesized Speech Using Multi-Task Learning With Spoofing Detection and Spoofing Type Classification](https://mlsanity.com/paper/2007.08267/) (2020)

## Built on by (most cited first, 10 of 10)

- [MOSPC: MOS Prediction Based on Pairwise Comparison](https://mlsanity.com/paper/2306.10493/) (2023)
- [Non-Intrusive Speech Intelligibility Prediction for Hearing Aids using Whisper and Metadata](https://mlsanity.com/paper/2309.09548/) (2023)
- [Investigating Content-Aware Neural Text-To-Speech MOS Prediction Using Prosodic and Linguistic Features](https://mlsanity.com/paper/2211.00342/) (2022)
- [Evaluation of Speech Representations for MOS prediction](https://mlsanity.com/paper/2306.09979/) (2023)
- [APG-MOS: Auditory Perception Guided-MOS Predictor for Synthetic Speech](https://mlsanity.com/paper/2504.20447/) (2025)
- [Multi-Task Pseudo-Label Learning for Non-Intrusive Speech Quality Assessment Model](https://mlsanity.com/paper/2308.09262/) (2023)
- [SVSNet+: Enhancing Speaker Voice Similarity Assessment Models with Representations from Speech Foundation Models](https://mlsanity.com/paper/2406.08445/) (2024)
- [Improving Perceptual Audio Aesthetic Assessment via Triplet Loss and Self-Supervised Embeddings](https://mlsanity.com/paper/2509.03292/) (2025)
- [SALF-MOS: Speaker Agnostic Latent Features Downsampled for MOS Prediction](https://mlsanity.com/paper/2506.02082/) (2025)
- [Unifying Model and Layer Fusion for Speech Foundation Models](https://mlsanity.com/paper/2511.08389/) (2025)

## Abstract (arXiv)

We participated in the mean opinion score (MOS) prediction challenge, 2022. This challenge aims to predict MOS scores of synthetic speech on two tracks, the main track and a more challenging sub-track: out-of-domain (OOD). To improve the accuracy of the predicted scores, we have explored several model fusion-related strategies and proposed a fused framework in which seven pretrained self-supervised learned (SSL) models have been engaged. These pretrained SSL models are derived from three ASR frameworks, including Wav2Vec, Hubert, and WavLM. For the OOD track, we followed the 7 SSL models selected on the main track and adopted a semi-supervised learning method to exploit the unlabeled data. According to the official analysis results, our system has achieved 1st rank in 6 out of 16 metrics and is one of the top 3 systems for 13 out of 16 metrics. Specifically, we have achieved the highest LCC, SRCC, and KTAU scores at the system level on main track, as well as the best performance on the LCC, SRCC, and KTAU evaluation metrics at the utterance level on OOD track. Compared with the basic SSL models, the prediction accuracy of the fused system has been largely improved, especially on OOD sub-track.