arXiv · 2306.09979
Evaluation of Speech Representations for MOS prediction
Abstract
In this paper, we evaluate feature extraction models for predicting speech quality. We also propose a model architecture to compare embeddings of supervised learning and self-supervised learning models with embeddings of speaker verification models to predict the metric MOS. Our experiments were performed on the VCC2018 dataset and a Brazilian-Portuguese dataset called BRSpeechMOS, which was created for this work. The results show that the Whisper model is appropriate in all scenarios: with both the VCC2018 and BRSpeech- MOS datasets. Among the supervised and self-supervised learning models using BRSpeechMOS, Whisper-Small achieved the best linear correlation of 0.6980, and the speaker verification model, SpeakerNet, had linear correlation of 0.6963. Using VCC2018, the best supervised and self-supervised learning model, Whisper-Large, achieved linear correlation of 0.7274, and the best model speaker verification, TitaNet, achieved a linear correlation of 0.6933. Although the results of the speaker verification models are slightly lower, the SpeakerNet model has only 5M parameters, making it suitable for real-time applications, and the TitaNet model produces an embedding of size 192, the smallest among all the evaluated models. The experiment results are reproducible with publicly available source-code1 .
Explore related subjects
Keep this discovery
Frederico S. Oliveira, Edresson Casanova, Arnaldo Cândido Júnior, Lucas R. S. Gris, Anderson S. Soares, Arlindo R. Galvão Filho. 2023-06-16. Evaluation of Speech Representations for MOS prediction. https://arxiv.org/abs/2306.09979
Cite the original work for its findings. Save a collection to share your selection of sources.