sp-uhh sgmse issues - Githubissues

sp-uhh / sgmse

Score-based Generative Models (Diffusion Models) for Speech Enhancement and Dereverberation

MIT License

491 stars 74 forks source link

issues

Newest

Newest Most commented Recently updated Oldest Least commented Least recently updated

Always eats up VRAM

#55 genevera closed 1 week ago
3
what is the test dir?

#54 FrankCasbert opened 1 month ago
1
which checkpoint can be applied to bandwidth expansion tasks?

#53 huhuqwaszxedc opened 2 months ago
0
enhancement.py error

#52 musend closed 3 months ago
1
unsupported gpu architecture 'compute_89'

#51 a897456 closed 3 months ago
1
White noise for 48kHz model

#50 isomov closed 4 months ago
1
Duplicate: parser.add_argument("--attn_resolutions", type=int, nargs='+', default=[])

#49 turian closed 4 months ago
3
tmp

#48 TakumiHirose11 closed 6 months ago
0
Wrong output shape in enhancement.py

#47 PavelPanjaya closed 6 months ago
3
ONNX conversion

#46 signal1234 opened 6 months ago
0
The enhanced.py can't generate the enhanced.wav

#45 Lieber0402 closed 4 months ago
2
Can this code be trained not by the ddp strategy?

#44 Lieber0402 closed 4 months ago
1
Data set dimension issues during training

#43 SevenXander closed 4 months ago
1
the results and training loss

#42 DingGitTemp closed 8 months ago
3
Questions about equation (10) in the paper

#41 WangHelin1997 closed 8 months ago
0
same noise for different sets

#40 LeiJ-7 closed 8 months ago
0
Training loss

#39 yangdongchao closed 4 months ago
1
Can not get good result with really bad dataset

#38 Vadim2S closed 4 months ago
1
Overloaded results. Dataset requirements?

#37 Vadim2S closed 4 months ago
1
nvrtc: error: invalid value for --gpu-architecture (-arch)

#36 szh404 closed 9 months ago
3
The data simulation is not reproducible..

#35 LiChenda closed 11 months ago
2
How to make the SGMSE in causal way

#34 Leaou-Bin opened 1 year ago
0
OSError: libstdc++.so.6: cannot open shared object file

#33 winjii closed 9 months ago
5
could not find the monitored key in the returned metrics

#32 joeoct93 closed 9 months ago
1
About the checkpoint

#31 yangdongchao closed 1 year ago
0
Why i got Nan when i train the ncsnpp?

#30 manmushanhe closed 4 months ago
3
Expose NCSN++ architecture args to CLI

#29 taltalim closed 1 year ago
0
Try to run on CPU

#28 SediRzm closed 4 months ago
1
Missing file: fused.so

#27 PengyvWANG closed 1 year ago
3
train with higher sampling rate

#26 taalua closed 4 months ago
1
Network conditioning

#25 philgzl opened 1 year ago
0
causality of the model

#24 XZWY closed 1 year ago
3
48k Checkpoint

#23 adeelabbas closed 1 year ago
1
Performance on RTX-3090

#22 gfh-dev closed 1 year ago
1
Changing the sample_rate of loaded audio file to 16000

#21 Aaryan369 closed 3 months ago
1
I have reduced the model size in half, but the RTF has doubled, what happend?

#20 Vuducbao913 opened 1 year ago
0
Reproducing results with VB-DMD testset

#19 k-sparrow closed 9 months ago
1
Where I can find the branch for "icassp_2023" ?

#18 utahboy3502 closed 1 year ago
1
The current PyTorch install supports CUDA capabilities sm_37 sm_50 sm_60 sm_70.

#17 zhourui-xihu closed 9 months ago
3
I can not reproduce enhanced audio for same noisy audio file. Question : sampler() ?

#16 ndisci closed 9 months ago
3
Questions on the evaluation on the VB-DMD dataset

#15 KeiKinn closed 2 years ago
2
What is the recommended hardware for evaluation?

#14 djm1176 closed 9 months ago
2
Question about performance gap between valid set & test set of VB-DMD dataset

#13 Kuray107 closed 2 years ago
4
RTF seems too high, any idea to improve it?

#12 NormonisPing closed 4 months ago
1
Runtime of inference and model size

#11 fakufaku closed 2 years ago
4
Bump protobuf from 3.19.4 to 3.19.5

#10 dependabot[bot] closed 9 months ago
1
How to use dereverbration checkpoint for inference?

#9 the6thsense closed 2 years ago
4
License for code in this repository

#8 stefan-baumann closed 2 years ago
2
Code for WSJ0-REVERB dataset reproduction

#7 stefan-baumann closed 2 years ago
4
OSError: CUDA_HOME environment variable is not set. Please set it to your CUDA install root.

#6 snufas closed 9 months ago
12