DTMSE: Masked Language Model Based on Discrete Tokens for Multi-Channel Speech Enhancement

In

Result Display

case 1
Reverberation+Noise (Ch. 0)
 Clean (Ch. 0)
 FaSNet
 EaBNet
McNet
 DeFT-AN
 CTFTSUNet
 DTMSE
case 2
Reverberation+Noise (Ch. 0)
 Clean (Ch. 0)
 FaSNet
 EaBNet
McNet
 DeFT-AN
 CTFTSUNet
 DTMSE
case 3
Reverberation+Noise (Ch. 0)
 Clean (Ch. 0)
 FaSNet
 EaBNet
McNet
 DeFT-AN
 CTFTSUNet
 DTMSE
case 4
Reverberation+Noise (Ch. 0)
 Clean (Ch. 0)
 FaSNet
 EaBNet
McNet
 DeFT-AN
 CTFTSUNet
 DTMSE
case 5
Reverberation+Noise (Ch. 0)
 Clean (Ch. 0)
 FaSNet
 EaBNet
McNet
 DeFT-AN
 CTFTSUNet
 DTMSE