@inproceedings{stahl2025distillation,
author = {Stahl, Benjamin and Gamper, Hannes},
title = {Distillation and Pruning for Scalable Self-Supervised Representation-Based Speech Quality Assessment},
booktitle = {2025 International Conference on Acoustics, Speech, and Signal Processing},
year = {2025},
month = {April},
abstract = {Distill-MOS is a compact and efficient speech quality assessment model learned from a larger speech quality assessment model based on wav2vec2.0 XLS-R embeddings. Model weights and inference code are available on GitHub.

Install Distill-MOS via

pip install distillmos

The work is described in the paper: "Distillation and Pruning for Scalable Self-Supervised Representation-Based Speech Quality Assessment".

Abstract:
In this paper, we investigate distillation and pruning methods to reduce model size for non-intrusive speech quality assessment based on self-supervised representations. Our experiments build on XLS-R-SQA, a speech quality assessment model using wav2vec 2.0 XLS-R embeddings. We retrain this model on a large compilation of mean opinion score datasets, encompassing over 100,000 labeled clips. For distillation, using this model as a teacher, we generate pseudo-labels on unlabeled degraded speech signals and train student models of varying sizes. For pruning, we use a data-driven strategy. While data-driven pruning performs better at larger model sizes, distillation on unlabeled data is more effective for smaller model sizes. Distillation can halve the gap between the baseline's correlation with ground-truth MOS labels and that of the XLS-R-based teacher model, while reducing model size by two orders of magnitude compared to the teacher model.},
url = {http://approjects.co.za/?big=en-us/research/publication/distill-mos-a-compact-speech-quality-assessment-model/},
}