Pith. sign in

A critical component in training enhancement models is the choice of loss function, which directly influences the quality and generaliza- tion of enhanced output [6, 7]

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

cs.SD 1

years

2025 1

verdicts

CONDITIONAL 1

roles

background 1

polarities

background 1

representative citing papers

Model as Loss: A Self-Consistent Training Paradigm

cs.SD · 2025-05-27 · conditional · novelty 6.0

Using the model's own encoder as a feature loss improves perceptual quality and iterative stability of a speech enhancement model compared with a WavLM-based loss.

citing papers explorer

Showing 1 of 1 citing paper.

  • Model as Loss: A Self-Consistent Training Paradigm cs.SD · 2025-05-27 · conditional · none · ref 1

    Using the model's own encoder as a feature loss improves perceptual quality and iterative stability of a speech enhancement model compared with a WavLM-based loss.