issues
search
Xiaobin-Rong
/
gtcrn
The official implementation of GTCRN, an ultra-lite speech enhancement model.
MIT License
202
stars
34
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
输入前的特征值问题
#44
ccf854
opened
4 hours ago
0
人声过度抑制问题
#43
hulucky1102
opened
5 days ago
0
作者你好，当人声中混有瞬态噪数（例如 严重的敲击声），这时模型会过度抑制人声来达到去除敲击声。请问这类情况可以怎样从数据或者损失函数得到改善么？
#42
hulucky1102
closed
5 days ago
0
Query on Sampling Rate Adaptation
#41
12oneway
closed
3 days ago
1
The training target is obtained by preserving the first 100 ms reflections
#40
zdongchen
closed
1 week ago
1
有没有计划支持AEC
#39
hfwanguanghui
opened
1 month ago
5
师兄，我有一个绝妙的想法
#38
yuyun2000
closed
1 month ago
0
端侧设备部署，onnx推理耗时20ms，是否正常以及还是否有优化手段
#37
lhbing-ai
opened
1 month ago
24
inference time
#36
xczhusuda
closed
1 month ago
1
Loss function - Magnitude compression values inside MSE Loss
#35
ercandogu-elevear
closed
2 months ago
1
模型流推理问题请教
#34
hulucky1102
closed
2 months ago
1
模型在rk3566上的速度依旧有点慢
#33
yuyun2000
closed
2 months ago
7
inference error
#32
yujinqiu
closed
2 months ago
2
GRNN representation rearrangement/group size
#31
ercandogu-elevear
closed
2 months ago
2
单频和扫频信号保留
#30
FisherDom
opened
2 months ago
1
关于该模型部署到嵌入式设备实现实时语音增强的问题
#29
canon1111k
closed
2 months ago
0
为什么使用 ERB 而不是 MFCC
#28
csu-roycheng
closed
2 months ago
1
关于输入特征会导致性能下降的问题
#27
TungyuYoung
opened
2 months ago
1
offline和online推理速度差异
#26
CaA23187
opened
3 months ago
1
关于流式推理GRU状态值传递的疑问
#25
IwraI
closed
3 months ago
1
训练效果和公开模型差异问题
#24
lflyme
opened
3 months ago
7
关于模型推理速度问题
#23
IwraI
closed
3 months ago
2
计算量是否可以进一步下探
#22
kaidachen
closed
3 months ago
1
hz2ERB的系数
#21
Fylll
opened
3 months ago
7
GTCRN实际上还是对幅度谱做mask？
#20
HuixiangH
opened
4 months ago
2
导出onnx的stream模型时可以优化一点点的两个方法
#19
SherryYu33
opened
4 months ago
9
DNS3 blind test set
#18
shenbuguanni
opened
4 months ago
4
数据制作脚本问题请教
#17
lflyme
closed
4 months ago
0
数据制作脚本问题请教
#16
lflyme
closed
4 months ago
0
数据制作脚本问题请教
#15
lflyme
opened
4 months ago
1
GTCRN的look-ahead是70ms吗
#14
Oliver0047
closed
4 weeks ago
2
对比模型的计算量问题
#13
jingxuan9862
opened
5 months ago
3
48k_training
#12
FisherDom
opened
6 months ago
5
Loss gets nan
#11
yezhangyinge
closed
5 months ago
14
请问，模型低信噪比场景听感难受问题
#10
gedebabin
opened
6 months ago
4
LIcense?
#9
fumminger
opened
6 months ago
1
Is this paper available for download online?
#8
pyg0629
opened
7 months ago
1
请问，如果想扩展一下，用计算量换模型的性能，应该朝哪个方向调参？
#7
zuowanbushiwo
closed
8 months ago
4
请问，训练代码在之后会开源吗
#6
wlf0322
opened
8 months ago
4
请教一下Temporal Recurrent Attention的作用
#5
JaySpeech
closed
8 months ago
2
提问，关于SFE模块使用nn.unfold的目的。
#4
TungyuYoung
opened
8 months ago
8
请问您排查到onnxsim.simplify导出时的问题所在了吗
#3
SherryYu33
closed
4 months ago
3
could you share the paper link? Thanks
#2
shenbuguanni
opened
9 months ago
12
can you share the paper, I can not find it in the internet
#1
ltcxjtu
closed
9 months ago
7