Partial mix-up on clean-degraded speech pairs plus contrastive loss produces frame-level embeddings that cluster by degradation type and improve detection and classification on in- and out-of-domain data.
The con- trastive loss also improved discrimination between degradation types and retrieval of the same types
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
eess.AS 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Speech Quality Embeddings for Improved Detection and Classification of Degradations in Speech Signals
Partial mix-up on clean-degraded speech pairs plus contrastive loss produces frame-level embeddings that cluster by degradation type and improve detection and classification on in- and out-of-domain data.