A multivariate Gaussian head on a wav2vec2 feature extractor jointly estimates MOS, noisiness, coloration, discontinuity, and loudness, matching point-estimate baselines and adding predicted score correlations.
Deepmos: Deep posterior mean-opinion-score of speech,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Multivariate Probabilistic Assessment of Speech Quality
A multivariate Gaussian head on a wav2vec2 feature extractor jointly estimates MOS, noisiness, coloration, discontinuity, and loudness, matching point-estimate baselines and adding predicted score correlations.