-
Notifications
You must be signed in to change notification settings - Fork 2.1k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
2021 Clarity Challenge recipe #4210
Conversation
This pull request is now in conflict :( |
� Conflicts: � espnet2/enh/layers/complex_utils.py � espnet2/enh/layers/dnn_beamformer.py
Codecov Report
@@ Coverage Diff @@
## master #4210 +/- ##
=======================================
Coverage 80.71% 80.71%
=======================================
Files 453 453
Lines 39578 39578
=======================================
+ Hits 31945 31946 +1
+ Misses 7633 7632 -1
Flags with carried forward coverage won't be shown. Click here to find out more.
📣 Codecov can now indicate which changes are the most critical in Pull Requests. Learn more |
Tests fails but probably not related to this PR |
egs2/clarity/enh_2021/local/data.sh
Outdated
|
||
. utils/parse_options.sh | ||
|
||
|
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Maybe add a check for the existence of sox
(as it is assumed to be available in local/prep_data.py
) like the following?
! command -v sox &>/dev/null && echo "sox: command not found" && exit 1;
"(Folder containing train, dev and metadata dirs)", | ||
) | ||
parser.add_argument( | ||
"--fs", |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Is this argument used?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
yes it is used now
import argparse | ||
import os | ||
import json | ||
from pathlib import Path |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Could you fix the import order to follow the alphabet order?
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
done
BTW, You may need to merge the latest master branch to fix the CI test error. |
Thanks for your suggestions ! Now it should be better. |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
LGTM now. Just two last comments:
- Could you add your recipe to egs2/README.md?
- Could you rename the recipe to
clarity21
to distinguish from the challenges in other years?
Can you upload a model to the HF hub and add a link to README.md? |
Sure, I'll look into how to do it |
Isn't it better to use a subfolder as I am using right now ? I am using |
This pull request is now in conflict :( |
It seems that espnet prefers to split them into different recipes, like chime4 and chime5. So it's better to also follow the convention here. |
� Conflicts: � egs2/README.md
https://huggingface.co/popcornell/clarity21_train_enh_beamformer_mvdr |
Thanks, @popcornell! |
This is an attempt to do a simple recipe for 2021 Clarity Challenge.
Currently missing: test set (I need to check if the oracle sources have been released for test set)
Also where do you set in recipes the path to the current root of the dataset ? What's the best practice ?