A causal speech enhancement model that combines self-supervised speech features with prediction of future quantized semantic tokens reaches 2.88 PESQ on VoiceBank+DEMAND, a 0.05 PESQ gain over not predicting semantics.
DCCRN: Deep complex convolution recurrent network for phase-aware speech enhancement,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
eess.AS 1years
2024 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Causal Speech Enhancement with Predicting Semantics based on Quantized Self-supervised Learning Features
A causal speech enhancement model that combines self-supervised speech features with prediction of future quantized semantic tokens reaches 2.88 PESQ on VoiceBank+DEMAND, a 0.05 PESQ gain over not predicting semantics.