Encoder layers in speech enhancement models maintain noise-invariant representations while decoder layers adapt strongly to degradation, with the pattern consistent across architectures and degradation types.
wav2vec 2.0: A framework for self- supervised learning of speech representations
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
method 1
citation-polarity summary
fields
eess.AS 2verdicts
UNVERDICTED 2roles
method 1polarities
use method 1representative citing papers
A hierarchical cross-modal fusion architecture with LoRA adapters predicts human perception of AI-dubbed clips at PCC > 0.75 after training on 12k Hindi-English examples and human MOS fine-tuning.
citing papers explorer
-
Where Does Speech Enhancement Adapt? Probing Study Under Controlled Degradation
Encoder layers in speech enhancement models maintain noise-invariant representations while decoder layers adapt strongly to degradation, with the pattern consistent across architectures and degradation types.
-
Can Hierarchical Cross-Modal Fusion Predict Human Perception of AI Dubbed Content?
A hierarchical cross-modal fusion architecture with LoRA adapters predicts human perception of AI-dubbed clips at PCC > 0.75 after training on 12k Hindi-English examples and human MOS fine-tuning.