Codestin Search App

Alexgichamba · 2024-02-29T14:37:29Z

What?

Add a spk1 recipe on the ASVspoof19 Logical Access ASV subset.

Why?

Provide diverse recipes for ESPnet-SPK, adding an SASV example.

Codecov Report

All modified and coverable lines are covered by tests ✅

Project coverage is 76.61%. Comparing base (c69d48e) to head (ed1691d).

Additional details and impacted files

@@            Coverage Diff             @@
##           master    #5687      +/-   ##
==========================================
+ Coverage   72.21%   76.61%   +4.39%     
==========================================
  Files         760      761       +1     
  Lines       69840    69880      +40     
==========================================
+ Hits        50435    53536    +3101     
+ Misses      19405    16344    -3061

Flag	Coverage Δ
test_configuration_espnet2	`∅ <ø> (∅)`
test_integration_espnet1	`62.92% <ø> (ø)`
test_integration_espnet2	`48.84% <ø> (?)`
test_integration_espnetez	`27.98% <ø> (ø)`
test_python_espnet1	`18.20% <ø> (ø)`
test_python_espnet2	`52.41% <ø> (ø)`
test_python_espnetez	`13.95% <ø> (ø)`
test_utils	`20.91% <ø> (ø)`

Flags with carried forward coverage won't be shown. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

Jungjee

Thank you for the work, Alex!
I left a few comments, but in general, it looks great!

Meanwhile, please make additional required updates to make all CI tests pass.

egs2/TEMPLATE/spk1/db.sh

egs2/asvspoof/spk1/local/asv_data_prep.py

Jungjee · 2024-03-25T02:07:42Z

egs2/asvspoof/spk1/local/asv_data_prep.py

+    ) as f_utt2spk, open(os.path.join(dst, "wav.scp"), "w") as f_wav:
+
+        for spk in spk2utt:
+            f_spk2utt.write(f"{spk} {' '.join(spk2utt[spk])}\n")


note that there's an utt2spk to spk2utt script within ESPnet.

Jungjee · 2024-03-25T02:11:14Z

egs2/asvspoof/spk1/local/cm_data_prep.py

@@ -0,0 +1,39 @@
+#!/usr/bin/env bash


[question] Where do we need this?
Do we need the train and the validation sets as well?

No. I did not use this. I thought perhaps someone using the recipe might want to do some in-domain finetuning. I could take it out if it is not needed.

egs2/asvspoof/spk1/local/data.sh

egs2/asvspoof/spk1/run.sh

Alexgichamba · 2024-03-25T05:30:10Z

Thank you for the work, Alex! I left a few comments, but in general, it looks great!

Meanwhile, please make additional required updates to make all CI tests pass.

Hi @Jungjee ,
Thanks for the review. I will address the comments.

for more information, see https://pre-commit.ci

Alexgichamba · 2024-03-27T01:46:05Z

Hi @Jungjee,
I have addressed the comments, except for two of them.

Should I remove the CM data prep? I had figured it might be useful later for in-domain finetuning
For the utt2spk/spk2utt, should I change it and use the script instead?
The CI checks that I am failing seem to be from install_espnet or test_python/test/espnet2/gan_tts. I am not quite sure how to make these pass

sw005320 · 2024-03-27T11:49:58Z

The CI checks that I am failing seem to be from install_espnet or test_python/test/espnet2/gan_tts. I am not quite sure how to make these pass

I reran the CI

Alexgichamba · 2024-03-27T13:09:38Z

The CI checks that I am failing seem to be from install_espnet or test_python/test/espnet2/gan_tts. I am not quite sure how to make these pass

I reran the CI

Thank you. Seems to be okay now

Jungjee · 2024-03-27T14:26:05Z

Thanks @Alexgichamba for the updates and your effort!!

Should I remove the CM data prep? I had figured it might be useful later for in-domain fine-tuning

I see. For me, it's okay to leave, but not so sure how @sw005320 would think about this.

For the utt2spk/spk2utt, should I change it and use the script instead?

Yes, it may give more consistency with other parts. But to me, the current status is also okay.

The CI checks that I am failing seem to be from install_espnet or test_python/test/espnet2/gan_tts. I am not quite sure how to make these pass

Now it's gone, thanks Shinji!

sw005320 · 2024-03-27T14:35:42Z

Thanks @Alexgichamba for the updates and your effort!!

Should I remove the CM data prep? I had figured it might be useful later for in-domain fine-tuning

I see. For me, it's okay to leave, but not so sure how @sw005320 would think about this.

We can leave it, then.

sw005320 · 2024-03-27T14:40:13Z

Thanks, @Alexgichamba
If you have some follow-up activities, please make a PR.

Alexgichamba · 2024-03-27T14:43:51Z

Thanks, @Alexgichamba If you have some follow-up activities, please make a PR.

Thanks @sw005320 and @Jungjee. I appreciate the support!

Alex Gichamba and others added 20 commits February 21, 2024 11:06

initial commit

1d1cf7b

add conf

20e99d3

add local files and run

8561e51

data prep

e045dfe

Merge branch 'espnet:master' into asvspoof19_spk

6591872

allow for train and adjust configs

2f881ff

add README

63aaa73

add missing prep file

68bfc27

fix unzip bug

4eead40

add mfa and xvec conf

659ba0a

Merge branch 'asvspoof19_spk' of https://github.com/Alexgichamba/espnet…

751930e

… into asvspoof19_spk

add results

8dd9554

fix README and update conf

b2c11e6

add eval for subprotocols

8618478

add rawnet3 subprotocol results

27b8eab

update SKA results

7113cb3

add ecapa results

79bf364

add results for mfa and xvec

fb1b1a4

Merge branch 'espnet:master' into asvspoof19_spk

2f49348

specify sasv-eer

e92f86d

mergify bot added ESPnet2 README labels Feb 29, 2024

[pre-commit.ci] auto fixes from pre-commit.com hooks

6377d1b

for more information, see https://pre-commit.ci

Alexgichamba marked this pull request as ready for review February 29, 2024 14:58

Merge branch 'master' into asvspoof19_spk

a9faf9f

sw005320 added the auto-merge Enable auto-merge label Feb 29, 2024

sw005320 added this to the v.202405 milestone Feb 29, 2024

sw005320 added SID Speaker identification/embedding and removed auto-merge Enable auto-merge labels Feb 29, 2024

Alexgichamba and others added 3 commits February 29, 2024 11:47

address test_shell errors

ea3a188

Merge branch 'master' into asvspoof19_spk

7456b1b

Merge branch 'master' into asvspoof19_spk

dce4a99

Jungjee reviewed Mar 25, 2024

View reviewed changes

Alexgichamba and others added 7 commits March 26, 2024 12:40

update corpus name

9ef7478

rm misplaced log

25fe639

add script description

0031706

Merge branch 'master' into asvspoof19_spk

55d5797

update corpus name in local/data sh

25d9ad0

add support for dev set

85a0877

[pre-commit.ci] auto fixes from pre-commit.com hooks

ed1691d

for more information, see https://pre-commit.ci

sw005320 merged commit 7265e7d into espnet:master Mar 27, 2024

Conversation

Alexgichamba commented Feb 29, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

What?

Why?

See also

Uh oh!

sw005320 commented Feb 29, 2024

Uh oh!

codecov bot commented Feb 29, 2024 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Codecov Report

Uh oh!

Jungjee left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Jungjee Mar 25, 2024

Choose a reason for hiding this comment

Uh oh!

Jungjee Mar 25, 2024

Choose a reason for hiding this comment

Uh oh!

Alexgichamba Mar 25, 2024

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Alexgichamba commented Mar 25, 2024

Uh oh!

Alexgichamba commented Mar 27, 2024

Uh oh!

sw005320 commented Mar 27, 2024

Uh oh!

Alexgichamba commented Mar 27, 2024

Uh oh!

Jungjee commented Mar 27, 2024

Uh oh!

sw005320 commented Mar 27, 2024

Uh oh!

sw005320 commented Mar 27, 2024

Uh oh!

Alexgichamba commented Mar 27, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

Alexgichamba commented Feb 29, 2024 •

edited

Loading

codecov bot commented Feb 29, 2024 •

edited

Loading

Jungjee left a comment •

edited

Loading