issues
search
aim-qmul
/
sdx23-aimless
Source Separation training codebase for the Sound Demixing Challenge 2023.
35
stars
0
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
feat: winning configurations
#51
yoyololicon
closed
1 year ago
0
fix: zero segment length
#50
yoyololicon
closed
1 year ago
0
Add clean LabelNoise logic
#49
mhrice
opened
1 year ago
2
feat: speech enhancement training
#48
yoyololicon
closed
1 year ago
1
dataset and loader for speech enhancement task
#47
yoyololicon
closed
1 year ago
0
train speech enhancement model for CDX
#46
yoyololicon
closed
1 year ago
0
Add dataset_split_and_mix script. Remove scripts dir being gitignored
#45
mhrice
closed
1 year ago
0
Fixing augmentations and enable pretrained HDemucs
#44
csteinmetz1
opened
1 year ago
1
Symbolic domain source separation with EnCodec model
#43
yoyololicon
opened
1 year ago
1
Base dataset
#42
mhrice
closed
1 year ago
0
feat: Demucs + Transformer-XL
#41
yoyololicon
closed
1 year ago
0
Runtime error of `RandomVolumeAutomation`
#40
yoyololicon
closed
1 year ago
0
feat: random panning
#39
yoyololicon
closed
1 year ago
0
Add class diagram in readme
#38
yoyololicon
opened
1 year ago
0
random stereo effects for dnr dataset
#37
yoyololicon
closed
1 year ago
0
feat: configurable callbacks using configs
#36
yoyololicon
closed
1 year ago
0
[WIP] implement DPT
#35
ben-hayes
opened
1 year ago
0
[cm] Update .gitignore
#34
christhetree
closed
1 year ago
0
Test dnr script on EECS servers
#33
yoyololicon
closed
1 year ago
0
baseline model on label noise dataset
#32
christhetree
closed
1 year ago
0
feat: aimless python package
#31
yoyololicon
closed
1 year ago
0
Add script to process speech, sfx, music into dnr dataset format
#30
mhrice
closed
1 year ago
1
[WIP] Integrate audio effects
#29
csteinmetz1
closed
1 year ago
6
Implement cinematic sound dataset mixing process (for non-DnR datasets)
#28
mhrice
closed
1 year ago
0
refactor: change config targets to set type
#27
yoyololicon
closed
1 year ago
0
feat: make data augmentation pipeline configurable
#26
yoyololicon
closed
1 year ago
0
Change `model.init_args.targets` type from list to set
#25
yoyololicon
closed
1 year ago
0
Make the training repo installable
#24
yoyololicon
closed
1 year ago
2
Lint on push and PR
#23
ben-hayes
closed
1 year ago
5
configurable data augmentation via yaml config
#22
yoyololicon
closed
1 year ago
0
mixture invariant training
#21
ben-hayes
opened
1 year ago
0
Handling source ordering
#20
csteinmetz1
closed
1 year ago
4
Fix issue with uploading mono files. Add missing aimless-logo-crop.png
#19
mhrice
opened
1 year ago
0
Run streamlit using user supplied checkpoint
#18
csteinmetz1
opened
1 year ago
0
Utility for inspecting model outputs
#17
yoyololicon
opened
1 year ago
0
feat: impulse responses augmentation class
#16
yoyololicon
closed
1 year ago
0
Basic streamlit webapp with pretrained demucs
#15
csteinmetz1
closed
1 year ago
2
Gather private datasets
#14
yoyololicon
opened
1 year ago
5
Model implementations
#13
yoyololicon
closed
1 year ago
1
Create base class for all dataset
#12
yoyololicon
closed
1 year ago
0
Impulse response augmentation
#11
yoyololicon
closed
1 year ago
0
audio effects augmentation
#10
yoyololicon
closed
1 year ago
1
Discussions on how to submit model for evaluation
#9
yoyololicon
closed
1 year ago
1
implement DPTNet
#8
yoyololicon
opened
1 year ago
1
implement DCUNet
#7
yoyololicon
opened
1 year ago
1
permutation invariant training
#6
yoyololicon
opened
1 year ago
5
use distribution object as an argument for sampling lufs values
#5
yoyololicon
opened
1 year ago
0
feat: DnR dataset for the cinematic track
#4
yoyololicon
closed
1 year ago
0
Add example youtube demxing pipeline
#3
mhrice
closed
1 year ago
0
feat: LimitAug
#2
yoyololicon
closed
1 year ago
0
Next