Investigating Training Objectives for Generative Speech Enhancement

Open paper
Benchmark Model Rank Results
speech-enhancement-on-ears-whamSchrödinger Bridge (PESQ loss)#1PESQ-WB: 3.09SI-SDR: 16.29ESTOI: 0.73SIGMOS: 3.18DNSMOS: 3.72…
speech-enhancement-on-demandSchrödinger bridge (PESQ loss)#3PESQ (wb): 3.70

Rank counts only results with a code link.