
Location
Badges
Activity
Challenge Categories
Challenges Entered
Audio Source Separation using AI
Latest submissions
See Allfailed | 219841 | ||
failed | 219840 | ||
graded | 219656 |
Music source separation of an audio signal into separate tracks for vocals, bass, drums, and other
Latest submissions
See Allfailed | 219841 | ||
failed | 219840 | ||
graded | 219656 |
Source separation of a cinematic audio track into dialogue, sound-effects and misc.
Latest submissions
Participant | Rating |
---|---|
![]() |
0 |
![]() |
0 |
Participant | Rating |
---|---|
![]() |
0 |
![]() |
0 |
![]() |
0 |
![]() |
0 |
![]() |
0 |
![]() |
0 |
![]() |
0 |
![]() |
0 |
![]() |
114 |
![]() |
0 |
![]() |
0 |
![]() |
0 |
![]() |
0 |
![]() |
0 |
![]() |
0 |
![]() |
0 |
|
Cinematic Sound Demixing Track - CDX'23


Secrets of success (MDX, leaderboard C)
About 2 years ago@JackZ @JusperLee @ouYi @XavierJ @MalcolmX
If someone will share how to get >9.900 on vocals, then I will share how to get this result:
9.851 bass
9.210 drums
6.603 other
9.168 vocals

Veni, vidi, vici
About 2 years agoIt looks a little strange that at the end of the challenge a participant from Korea University “junicus” gets 3rd and 2nd place on the A and B leaderboards on the first attempt. Another leader, the “kuielab” team is also from Korea University. I hope that AIcrowd Team will carefully check the code of these participants for borrowing (private sharing) after the challenge ends.

Slow downloading of musdb18hq
About 2 years agoMy training pipeline for leaderboard A and B starts with downloading of dataset from https://sdx-2023-data-bucket.s3.amazonaws.com. Dear organizers, could you upload full version of musdb18hq to this cloud? Download speed from zenodo is about 400 KB/s.

Feedback and suggestion
About 2 years ago@mohanty
It would be more convenient if the submission counter resets at 00:00 UTC

Reasons of submission failure
About 2 years ago@dipam
Also please rerun #214696 and #214694 (“Evaluation timed out”)

Reasons of submission failure
About 2 years ago#214686
#214701
Can you please rerun these submissions?
(“Scoring failed. Please contact the admins”)
Today, due to my mistake, 4 submissions failed with “Prediction too slow”. They ran simultaneously. Could this have somehow overloaded the system and caused scoring to fail?


Secrets of success (MDX, leaderboard C)
About 2 years agoCan anyone share any ideas on how the top 3 teams on the “C” leaderboard got such high SDR on “vocals” and “other”?



Reasons of submission failure
Over 2 years ago#210459
Demixing got stuck (2/2 - 0.689x, 6/27 - 0.692x), then I got “Evaluation timed out”

Reasons of submission failure
Over 2 years ago#210461
After successful demixing (2/2 - 1.73x, 27/27 - 1.749x) I got the message: “Evaluation timed out”

Reasons of submission failure
Over 2 years ago#210438
After demixing I got the message: “Scoring failed. Please contact the admins.”

What baselines are coming up?
Over 2 years agoThanks. Can you please post the entire presentation from town hall?

What baselines are coming up?
Over 2 years agoWhat baselines for MDX (A,B,C) and CDX (A,B) will be published during the challenge?


Reasons of submission failure
Over 2 years agoToday I successfully resubmitted code from failed #210197 (2/2 - 0.771x, 3/27 - 0.769x) as #210246 (2/2 - 0.757x, 27/27 - 0.749x)

Reasons of submission failure
Over 2 years agoMy understanding is that the speed of my model is constant.
#210197 (2/2 - 0.771x, 3/27 - 0.769x) may have fallen due to low speed.
It is less likely that #210059 fell due to low speed (2/2 - 0.926x, 9/27 - 0.913x).
#210078 and #210117 are identical. #210078 returned “Evaluation timed out” during validation phase. #210117 was evaluated quickly and successfully (2/2 - 3.369x, 27/27 - 3.479x).
I think it’s better to turn on the second counter instead of looking for the causes of rare crashes.
Secrets of success (MDX, leaderboard B)
About 2 years agoHello everyone. I’m currently in 4th place on the leaderboard B. I’m using demucs baseline and another model trained from scratch for vocals. It would be funny if someone told me how to improve my result by 0.24 to get into the prizes