Speech Denoising on VB-Demand (test)
3.27PESQKu et al.
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Ku et al.Sampling rate=16kHz, Adaptation=Task-specific finetuning2025.09 | 3.27 | 4.41 | 88 | 19.1 | |
| SpeechFlowSampling rate=16kHz, Adaptation=Task-specific finetuning2025.09 | 3.13 | — | 87 | — | |
| VoiceBridgeSampling rate=48kHz, Adaptation=No task-specific adaptation (General model)2025.09 | 2.83 | 4.39 | 79 | 4.8 |