SSL-MOS pretrained model

Pretrained model from our ICASSP 2022 paper:
"Generalization Ability of MOS Prediction Networks"
Erica Cooper, Wen-Chin Huang, Tomoki Toda, Junichi Yamagishi
Presented at ICASSP 2022
https://doi.org/10.1109/ICASSP43922.2022.9746395
Please cite this paper if you use this pretrained model.

This model is meant to be used with the following code on github:
https://github.com/nii-yamagishilab/mos-finetune-ssl
Please see that codebase's README for more information on how to use this model.

This model was the baseline "B01" in the VoiceMOS Challenge 2022.
https://voicemos-challenge-2022.github.io
It is a wav2vec2-base model from Fairseq:
https://github.com/facebookresearch/fairseq/blob/main/examples/wav2vec/README.md
The base model was fine-tuned for the MOS prediction task on the BVCC dataset:
https://doi.org/10.5281/zenodo.6572573