HyperAI
Command Palette
Search for a command to run...
语音分离与增强
从混合音频中分离出独立语音信号（分离）并通过去除噪声提升语音质量（增强）。
排行榜
共计 39 个模型
基准测试
VoiceBank + DEMAND
CMGAN
Deep Noise Suppression (DNS) Challenge
Conv-TasNet-SNR
EARS-WHAM
Schrödinger bridge (PESQ loss)
WHAMR!
Conv-TasNet DAE
CHiME-3
Inter-Channel Conv-TasNet
VCTK
TIMIT
LRS2
TDFNet-small
iEEG Clinical Dataset
IEEE
DNS challenge
3rd AVSEC dataset
LRS3+DNS
LibriTTS Test-clean
LibriSpeech test-clean
LibriSpeech
DNS3
VoiceBank+DEMAND
VBD
Speech Separation
Speech-Sound Event Separation
TCD-TIMIT corpus (mixed-speech)
WSJ0-SI 84+NOISEX-92
TIMIT+MUSAN
WSJ0-2mix-extr
WHAMR_ext
Conv-TasNet DAE
unseen speaker test set
VibraVox (forehead accelerometer)
WHAM!
SepFormer
VibraVox (throat microphone)
VibraVox (temple vibration pickup)
VibraVox (soft in-ear microphone)
VibraVox (rigid in-ear microphone)
EARS-Reverb
SGMSE+
AudioMarkBench
AVDnR
AVED
AVSpeech + AudioSet
CHiME-4
Cocktail Party
DEMAND
Wave-U-Net
DnRv2
DnRv3
spatialized WSJCAM0
EasyCom
MaxDI (Baseline)
GRID corpus (mixed-speech)
Audio-Visual concat-ref
LibriTTS Test-other
LJ Speech
MLS
MusicCaps
Sound Event Separation
SoundSpaces-Speech
spatialized DNS challenge
DeFT-AN