Showing eess.ASShow all
3 papers · 1 filter
eess.AS2024
The PESQetarian: On the Relevance of Goodhart's Law for Speech Enhancement
Danilo de Oliveira, Simon Welker, Julius Richter +1
To obtain improved speech enhancement models, researchers often focus on increasing performance according to specific instrumental metrics. However, when the same metric is used in…
eess.AS2023
Distilling HuBERT with LSTMs via Decoupled Knowledge Distillation
Danilo de Oliveira, Timo Gerkmann
Much research effort is being applied to the task of compressing the knowledge of self-supervised models, which are powerful, yet large and memory consuming. In this work, we show…
eess.AS2023
On the Behavior of Intrusive and Non-intrusive Speech Enhancement Metrics in Predictive and Generative Settings
Danilo de Oliveira, Julius Richter, Jean-Marie Lemercier +2
Since its inception, the field of deep speech enhancement has been dominated by predictive (discriminative) approaches, such as spectral mapping or masking. Recently, however, nove…