Codestin Search App

popcornell · 2024-01-28T03:26:20Z

The idea is to also update CHiME-7 DASR baseline to work for CHiME-8 DASR challenge.
We can then use it for the new challenge paper and also some participants would like maybe to use ESPNet.

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

# Conflicts: # egs2/chime8_task1/asr1/local/da_wer_scoring.py

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

sw005320 · 2024-02-12T12:24:41Z

Please let us know if this PR is ready for review.
@simpleoier, can you review this PR?

popcornell · 2024-02-12T14:18:57Z

Will do, I am training a new model based on e-branchformer now.
Hopefully will improve the performance.

popcornell · 2024-02-12T14:20:27Z

egs2/chime8_task1/asr1/run.sh

+    asr_tag="$(basename "${asr_config}" .yaml)_raw"
+    asr_exp="exp/asr_${asr_tag}"
+  fi
+  inference_tag="$(basename "${inference_config}" .yaml)"


This is currently failing BTW.
basically due to bash overriding the learning rate and batch size the name of the trained model will be different than the name of the YAML file.
Is there a way to set the name for the ASR model folder ? @simpleoier

As we discussed, this is due to the difference of asr_tag here in this file and asr.sh.

simpleoier

You may add chime7 and chime8 in egs2/README.md.

simpleoier · 2024-02-12T18:40:30Z

egs2/chime8_task1/asr1/baseline_logs/RESULTS.md

+- espnet version: `espnet 202301`
+- pytorch version: `pytorch 1.13.1`
+- Git hash: `89ebca463c544dfaa19e5f76ad5f615f473f6957`
+  - Commit date: `Tue Mar 7 04:02:43 2023 +0000`


you may put the pretrained checkpoint here.

Please do that in the follow-up PR then.

We don't need that it is in HF the model

simpleoier · 2024-02-12T18:42:14Z

egs2/chime8_task1/asr1/baseline_logs/inference.log

Is it possible to rename this result file (from the script)? *.log is not a good name sometimes.
In addition, there are some information related to your own cluster.

Yes we can rename into txt

simpleoier · 2024-02-12T18:42:54Z

egs2/chime8_task1/asr1/conf/train_lm.yaml

Was LM used? Maybe this file can be excluded.

No it was not used

I prefer to leave it I don't have time to retest everything and find out if it is okay to remove

simpleoier · 2024-02-12T18:47:25Z

egs2/chime8_task1/asr1/run.sh

+    asr_tag="$(basename "${asr_config}" .yaml)_raw"
+    asr_exp="exp/asr_${asr_tag}"
+  fi
+  inference_tag="$(basename "${inference_config}" .yaml)"


As we discussed, this is due to the difference of asr_tag here in this file and asr.sh.

simpleoier · 2024-02-12T18:47:54Z

egs2/chime8_task1/diar_asr1/README.md

It is still chime-7 written in this file.

I moved README.md and changed it

popcornell · 2024-02-15T14:21:09Z

This is ready.
I added the results

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

simpleoier

I left some minor comments. I think this PR looks good to me. Thanks!

simpleoier · 2024-02-15T17:44:41Z

egs2/chime8_task1/asr1/README.md

+
+Such baseline system would rank third on dev set based on the rules of the past CHiME-6 Challenge
+on Track 1 (unconstrained LM).
+Results on the evaluation set will be released after the end of the CHiME-7 DASR Challenge. <br>


Should this description be updated?

Yes thanks ! i removed this file actually

simpleoier · 2024-02-15T17:48:12Z

egs2/chime8_task1/asr1/local/da_wer_scoring.py

+if __name__ == "__main__":
+    parser = argparse.ArgumentParser(
+        "This script is used for scoring according to the procedure outlined in"
+        " CHiME-7 DASR challenge website"


It's in chime-8 folder and the website points to latest challenge. But it should be fine.

It is okay this is not used anymore for scoring

but people may want still to use it

popcornell · 2024-02-15T18:40:30Z

@sw005320 let's merge this

popcornell · 2024-02-15T18:47:03Z

I have improved a bit the results:

+-----+--------------+--------------+----------+----------+--------------+-------------+-----------------+------------------+------------------+------------------+
|     | session_id   |   error_rate |   errors |   length |   insertions |   deletions |   substitutions |   missed_speaker |   falarm_speaker |   scored_speaker |
|-----+--------------+--------------+----------+----------+--------------+-------------+-----------------+------------------+------------------+------------------|
| dev | chime6       |     0.825381 |    52070 |    63086 |        12747 |       29466 |            9857 |                0 |                5 |                8 |
| dev | mixer6       |     0.287729 |    26621 |    92521 |         4882 |        8809 |           12930 |                0 |               24 |               70 |
| dev | dipco        |     0.674161 |    11574 |    17168 |         3066 |        5563 |            2945 |                0 |                2 |                8 |
| dev | notsofar1    |     0.508768 |    90660 |   178195 |        14872 |       55195 |           20593 |              105 |                7 |              592 |
+-----+--------------+--------------+----------+----------+--------------+-------------+-----------------+------------------+------------------+------------------+
###############################################################################
### Macro-Averaged tcpWER for across all Scenario (Ranking Metric) ############
###############################################################################
+-----+--------------+
|     |   error_rate |
|-----+--------------|
| dev |      0.57401 |
+-----+--------------+

sw005320 · 2024-02-15T19:09:34Z

egs2/chime8_task1/asr1/baseline_logs/inference.log

Is it based on the markdown?

Thanks Shinji, these were there by error. I removed them.
I need to update here the logs with oracle diarization (currently running).

sw005320 · 2024-02-15T19:18:20Z

egs2/chime8_task1/asr1/local/generate_chime6_data.sh

Will you change it to simply download chime6 from openslr in the future?

I can remove this.
The download is done via chime-utils now

sw005320 · 2024-02-15T19:29:18Z

egs2/chime8_task1/diar_asr1/baseline_logs/tcpwer_scoring.log

Should we remove or should we keep it?

I think it is better to keep the scoring logs

sw005320 · 2024-02-15T19:53:52Z

egs2/chime8_task1/diar_asr1/baseline_logs/cpwer_scoring.log

Do you need this?

Yes it is better to report also cpWER

popcornell · 2024-02-15T22:34:10Z

I removed the files @simpleoier merge it as it is approved

sw005320 · 2024-02-16T00:02:57Z

The PR passed major CI tests. So, I merged it.
Thanks @popcornell!

added c8dasr recipe based on c7dasr one

7722f69

mergify bot added ESPnet2 README labels Jan 28, 2024

popcornell and others added 8 commits February 11, 2024 19:45

added new dataprep

87dac72

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

4b995d8

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

added c8dasr recipe based on c7dasr one

aa2fb80

added new dataprep

4f904fb

added e-branchformer

efb5f3d

removed old logs

21f5061

added diarization subdir

00b4262

Merge remote-tracking branch 'popcornell/c8dasr'

f42718c

# Conflicts: # egs2/chime8_task1/asr1/local/da_wer_scoring.py

mergify bot added ESPnet1 ASR Automatic speech recogntion TTS Text-to-speech Documentation CI Travis, Circle CI, etc Installation labels Feb 12, 2024

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

bd3180f

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

popcornell changed the base branch from master to multilingual February 12, 2024 14:17

popcornell changed the base branch from multilingual to master February 12, 2024 14:18

popcornell commented Feb 12, 2024

View reviewed changes

simpleoier reviewed Feb 12, 2024

View reviewed changes

popcornell added 6 commits February 13, 2024 18:25

fixed e branchformer

6ad92ec

apply normalization only on training of the ASR model

8cd9ed1

expanded data augmentation to include dipco and notsofar1

70db166

added top-k diarization.

67b1285

topk and sc selection

3c78ee4

sc selection

d125dc9

popcornell added 5 commits February 15, 2024 09:20

changes in data.sh

b1cb2f1

dprep of segmentation improved

cd7debc

add results

be71df7

added results

7d2e361

Merge branch 'master' into c8dasr

45230ef

pre-commit-ci bot and others added 6 commits February 15, 2024 14:21

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

ed788f5

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

added c8dasr recipe based on c7dasr one

c416ba8

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

b3ce937

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

changed threshold clust

1c5b3ab

address shellcheck

5e0fcaf

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

c2b0520

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

simpleoier approved these changes Feb 15, 2024

View reviewed changes

remove readme

97f85f3

updated results

8ccb7e6

sw005320 reviewed Feb 15, 2024

View reviewed changes

removed unnecessary files

476007e

sw005320 reviewed Feb 15, 2024

View reviewed changes

sw005320 added the auto-merge Enable auto-merge label Feb 15, 2024

sw005320 added this to the v.202405 milestone Feb 15, 2024

sw005320 approved these changes Feb 15, 2024

View reviewed changes

removed unnecessary files

3ad06d5

sw005320 added Recipe and removed TTS Text-to-speech ESPnet1 labels Feb 15, 2024

sw005320 merged commit 7ab5e42 into espnet:master Feb 16, 2024

Conversation

popcornell commented Jan 28, 2024

Uh oh!

sw005320 commented Feb 12, 2024

Uh oh!

popcornell commented Feb 12, 2024

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

simpleoier left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

popcornell commented Feb 15, 2024

Uh oh!

simpleoier left a comment

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

popcornell commented Feb 15, 2024

Uh oh!

popcornell commented Feb 15, 2024

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Choose a reason for hiding this comment

Uh oh!

popcornell commented Feb 15, 2024

Uh oh!

sw005320 commented Feb 16, 2024

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone