Replacing early-reflected speech with time-shifted anechoic clean speech as the training target, combined with a two-stage distortion-perception framework, yields state-of-the-art universal speech enhancement.
A two-stage training framework for joint speech compression and en- hancement.arXiv preprint arXiv:2309.04132,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.SD 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Rethinking Training Targets, Architectures and Data Quality for Universal Speech Enhancement
Replacing early-reflected speech with time-shifted anechoic clean speech as the training target, combined with a two-stage distortion-perception framework, yields state-of-the-art universal speech enhancement.