Closed shahryarghayoor closed 2 years ago
Hi Shahryar,
Thank you very much for your interest in our work! I saw that the command line you provide, i.e. "[ -d data/spec/${name} ] cp -r data/${name} data/spec/${name}", is not exactly the command in the line 42, Step 1, "extract_feats.sh". It should be "[ -d data/spec/${name} ] || cp -r data/${name} data/spec/${name} || exit 1". Please revise the command and see if the error still exists. Thank you!
Hi Shahryar,
Thank you very much for your interest in our work! I saw that the command line you provide, i.e. "[ -d data/spec/${name} ] cp -r data/${name} data/spec/${name}", is not exactly the command in the line 42, Step 1, "extract_feats.sh". It should be "[ -d data/spec/${name} ] || cp -r data/${name} data/spec/${name} || exit 1". Please revise the command and see if the error still exists. Thank you!
Thank you so much for answering actually it went well untill step 3. there is a massage here :
Processing PA_train spec
copy-feats --compress=true ark:- ark,scp:data/spec/PA_train/feats_slicing.ark,data/spec/PA_train/feats_slicing.scp
LOG (copy-feats[5.5.985~1-94f37]:main():copy-feats.cc:143) Copied 113676 feature matrices.
Processing PA_train cqt
[Errno 2] No such file or directory: 'data/cqt/PA_train/feats.scp'
copy-feats --compress=true ark:- ark,scp:data/cqt/PA_train/feats_slicing.ark,data/cqt/PA_train/feats_slicing.scp
LOG (copy-feats[5.5.985~1-94f37]:main():copy-feats.cc:143) Copied 0 feature matrices.
Exception in thread Thread-1:
Traceback (most recent call last):
File "/home/sh/anaconda3/lib/python3.8/threading.py", line 932, in _bootstrap_inner
self.run()
File "/home/sh/anaconda3/lib/python3.8/threading.py", line 870, in run
self._target(*self._args, **self._kwargs)
File "/home/sh/.local/lib/python3.8/site-packages/kaldi_io/kaldi_io.py", line 97, in cleanup
raise SubprocessFailed('cmd %s returned %d !' % (cmd,ret))
kaldi_io.kaldi_io.SubprocessFailed: cmd copy-feats --compress=true ark:- ark,scp:data/cqt/PA_train/feats_slicing.ark,data/cqt/PA_train/feats_slicing.scp returned 1 !
Traceback (most recent call last):
File "feats_extraction/convertID2index.py", line 194, in
where did I make a mistake? can you tell me please?
Hi Shahryar,
Thank you so much for pointing out this error. This could be a bug in my code. Actually, the missing utt2systemID
file is identical to that in the data/PA_train
directory, so you can copy the file from one to another. This also applies to the PA_dev
, PA_eval
, LA_train
, LA_dev
and LA_eval
.
I revised the extract_feats.sh
script, and please see Line 52 to 55. Thank you!
Hi Shahryar,
Thank you so much for pointing out this error. This could be a bug in my code. Actually, the missing
utt2systemID
file is identical to that in thedata/PA_train
directory, so you can copy the file from one to another. This also applies to thePA_dev
,PA_eval
,LA_train
,LA_dev
andLA_eval
. I revised theextract_feats.sh
script, and please see Line 52 to 55. Thank you!
Well thank you for your answer again. and it wokred as you mentioned.
actually I think feature extraction is done completely by codes. But when I start the evaluaton step "start.sh" part, this error code comes and seems "utt2index" was not created:
Traceback (most recent call last):
File "train.py", line 172, in
How can I create this file to avoid this problem?
I guess you may skip Step 3 in the extract_feats.sh
file. This step is necessary before training and evaluating our models, and it will create the utt2index
file automatically. Thank you!
I guess you may skip Step 3 in the
extract_feats.sh
file. This step is necessary before training and evaluating our models, and it will create theutt2index
file automatically. Thank you!
Thank you and it worked fine because of your help. I came to step 4 where lfcc is being extracted. at this step I could extract lfcc fils as .txt file but stuck at line 12 of extaction where it should be converted to .ark and .scp files. unfortunatly this error occurs:
/home/sh/Downloads/Compressed/ASV-anti-spoofing-with-Res2Net/baseline/write_feature_kaldi_LA_LFCC.sh: line 12: copy-feats: command not found
this is because the output is a .txt file in lfcc folder. I am not sure if I am ready to go to the evaluation process. I started the "start.sh" and it started "Start training" stage.
It seems that you need to install Kaldi and add the Kaldi path to the system before running the scripts.
Hello dear Dr. Xu Li
My name is Shahryar Ghayour an I am working on your paper ''ASV-anti-spoofing-with-Res2Net''.
I have a question if you can help me please.
I have installed kaldi and made changes to your ''extract_feats.sh '' file and added my directory to the dataset. the question is when I put the file in terminal on linux there is an error in step 1.
when it reaches to this line : [ -d data/spec/${name} ] cp -r data/${name} data/spec/${name}
an error comes which is [: missing `]'
how can I make it work?
can you please tell me?
Thank you so much