Comments (13)
Hi, I recommend you to set "debug=True" in build_spk_models.py and re-run the code.
It will show you whether Kaldi runs normally or not on each stage, and you can check the .log file at its instruction.
from fakebob.
I guess your issue is due to certain stage of Kaldi (extract MFCC, VAD, extract ivector, etc) fails, so scores file is empty.
from fakebob.
Thanks.
I turned on debug mode.
Looks like something is wrong with the cut. Trying to see why.
BTW, I am using mac os.
LOG (ivector-plda-scoring[5.5.708~2-fd88b]:main():ivector-plda-scoring.cc:147) Read 4716 test iVectors.
LOG (ivector-plda-scoring[5.5.708~2-fd88b]:main():ivector-plda-scoring.cc:150) Average renormalization scale on test iVectors was 0.983834
cut: illegal option -- -
usage: cut -b list [-n] [file ...]
cut -c list [file ...]
cut -f list [-s] [-d delim] [file ...]
LOG (ivector-plda-scoring[5.5.708~2-fd88b]:main():ivector-plda-scoring.cc:217) Processed 0 trials, 0 had errors.
WARNING (ivector-plda-scoring[5.5.708~2-fd88b]:Close():kaldi-io.cc:515) Pipe cat /Users/user1/FAKEBOB/ivector-build-model-iv/trials | cut -d\ --fields=1,2 | had nonzero return status 256
/Users/user1/FAKEBOB/ivector_PLDA_kaldiHelper.py:285: UserWarning: loadtxt: Empty input file: "/Users/user1/FAKEBOB/ivector-build-model-iv/scores"
scores_mat = np.loadtxt(scores_file, dtype=str)
Traceback (most recent call last):
File "build_spk_models.py", line 112, in
iv_helper.score_existing(audio_path_list, enroll_utt_id, spk_id_list=spk_id_list,
File "/Users/user1/FAKEBOB/ivector_PLDA_kaldiHelper.py", line 334, in score_existing
score_array = self.resolve_score()
File "/Users/user1/FAKEBOB/ivector_PLDA_kaldiHelper.py", line 290, in resolve_score
from fakebob.
OK.. so it works with Ubuntu, MacOS is using different cut syntax.
from fakebob.
I fixed several mac syntax bugs. But I am having another error report. I am trying to stick with the MacOS because it is most convenient machine for me to work with. Can you help give some clue on this error?
Thanks
`
--- obtaining gmm identity by updating ubm via MAP ---
--- obtaining gmm identity by updating ubm via MAP done ---
--- calculate z-norm mean, z-norm std ---
fix_data_dir.sh: kept all 4711 utterances.
fix_data_dir.sh: old files are kept in /Users/user1/FAKEBOB/audio-build-model-gmm/.backup
/Users/user1/FAKEBOB/pre-models/steps/make_mfcc.sh --write-utt2num-frames true --mfcc-config /Users/user1/FAKEBOB/pre-models/conf/mfcc.conf --nj 6 --cmd $train_cmd /Users/user1/FAKEBOB/audio-build-model-gmm /Users/user1/FAKEBOB/log-build-model-gmm /Users/user1/FAKEBOB/mfcc-build-model-gmm
utils/validate_data_dir.sh: Successfully validated data-directory /Users/user1/FAKEBOB/audio-build-model-gmm
/Users/user1/FAKEBOB/pre-models/steps/make_mfcc.sh: [info]: no segments file exists: assuming wav.scp indexed by utterance.
/Users/user1/FAKEBOB/pre-models/steps/make_mfcc.sh: Succeeded creating MFCC features for audio-build-model-gmm
/Users/user1/user1/pre-models/sid/compute_vad_decision.sh --nj 6 --cmd $train_cmd --vad-config /Users/user1/FAKEBOB/pre-models/conf/vad.conf /Users/user1/FAKEBOB/audio-build-model-gmm /Users/user1/FAKEBOB/log-build-model-gmm /Users/user1/FAKEBOB/mfcc-build-model-gmm
Created VAD output for audio-build-model-gmm
run.pl: 6 / 6 failed, log is in /Users/user1/FAKEBOB/log-build-model-gmm/get_frame_likes.*.log
Traceback (most recent call last):
File "build_spk_models.py", line 261, in
score_array = gmm_helper.score_existing(model_path_list, z_norm_utt_path, n_jobs=n_jobs, debug=debug)
File "/Users/user1/FAKEBOB/gmm_ubm_kaldiHelper.py", line 265, in score_existing
self.get_frames_likes(model_path_list, n_jobs=n_jobs, debug=debug)
File "/Users/user1/FAKEBOB/gmm_ubm_kaldiHelper.py", line 228, in get_frames_likes
with open(job_score_file, "r") as reader:
FileNotFoundError: [Errno 2] No such file or directory: '/Users/user1/FAKEBOB/score-build-model-gmm/score.1'`
from fakebob.
Hi, could you provide the log file /Users/user1/FAKEBOB/log-build-model-gmm/get_frame_likes.1.log ? It may help to debug.
from fakebob.
The line "run.pl: 6 / 6 failed, log is in /Users/user1/FAKEBOB/log-build-model-gmm/get_frame_likes.*.log" show that the GMM get_frame_likes stage failed. Need the log file to find the reason.
from fakebob.
this is get_frame_likes.1 file
# gmm-global-get-frame-likes --average=true /Users/user1/FAKEBOB/model/1580-identity.gmm "ark,s,cs:add-deltas --delta-window=3 --delta-order=2 scp:/Users/user1/FAKEBOB/audio-build-model-gmm/split6/1/feats.scp ark:- | apply-cmvn-sliding --norm-vars=false --center=true --cmn-window=300 ark:- ark:- | select-voiced-frames ark:- scp,s,cs:/Users/user1/FAKEBOB/audio-build-model-gmm/split6/1/vad.scp ark:- |" ark,t:/Users/user1/FAKEBOB/score-build-model-gmm/score.1 || exit 1;
# Started at Wed Jun 17 16:39:23 PDT 2020
# gmm-global-get-frame-likes --average=true /Users/user1/FAKEBOB/model/1580-identity.gmm 'ark,s,cs:add-deltas --delta-window=3 --delta-order=2 scp:/Users/user1/FAKEBOB/audio-build-model-gmm/split6/1/feats.scp ark:- | apply-cmvn-sliding --norm-vars=false --center=true --cmn-window=300 ark:- ark:- | select-voiced-frames ark:- scp,s,cs:/Users/user1/FAKEBOB/audio-build-model-gmm/split6/1/vad.scp ark:- |' ark,t:/Users/user1/FAKEBOB/score-build-model-gmm/score.1 ERROR (gmm-global-get-frame-likes[5.5.708~2-fd88b]:Input():kaldi-io.cc:756) Error opening input stream /Users/user1/FAKEBOB/model/1580-identity.gmm
[ Stack-Trace: ]
0 libkaldi-base.dylib 0x000000010b70292f kaldi::KaldiGetStackTrace() + 63
1 libkaldi-base.dylib 0x000000010b7026a2 kaldi::MessageLogger::LogMessage() const + 354
2 libkaldi-util.dylib 0x000000010c16d138 kaldi::MessageLogger::LogAndThrow::operator=(kaldi::MessageLogger const&) + 24
3 libkaldi-util.dylib 0x000000010c17716b kaldi::Input::Input(std::__1::basic_string<char, std::__1::char_traits, std::__1::allocator > const&, bool*) + 139
4 gmm-global-get-frame-likes 0x0000000109975d28 main + 392
5 libdyld.dylib 0x00007fff63eb87fd start + 1
6 ??? 0x0000000000000005 0x0 + 5
kaldi::KaldiFatalError# Accounting: time=1 threads=1
# Ended (code 1) at Wed Jun 17 16:39:24 PDT 2020, elapsed time 1 seconds
Also there is no 1580-identity.gmm under /FAKEBOB/model/
only 5 *.iv files exist
from fakebob.
It seems that the MAP stage failed, so there is no 1580-identity.gmm.
To figure out why it failed, please modify Line 208 and Line 218 of build_spk_models.py to p = subprocess.Popen(args) if debug else subprocess.Popen(args, stdout=subprocess.DEVNULL, stderr=subprocess.DEVNULL). I will also update this too.
from fakebob.
I guess the reason is that your gmm-global-est-map.cc has not been compiled to executable file, as step 2 instructs.
Type which gmm-global-est-map in terminal for quick check.
from fakebob.
Oh, it is my mistake.
In step 2, you should edit the file
Sorry for that. Please return to step 2 and rebuild kaldi. It should work well after rebuild.
from fakebob.
README has been updated.
from fakebob.
Thanks a lot. This step is working now.
from fakebob.
Related Issues (16)
- Something wrong in 'build_spk_models.py' or 'pre-models' HOT 2
- Is gpu used in Generate adversarial voices HOT 1
- Failed building Docker: Unable to establish SSL connection HOT 7
- New change on loss function HOT 2
- test.py: No such file or directory: 'model/1580.iv' HOT 8
- No such file or directory: 'ivector-plda-scoring' HOT 1
- No such file or directory: 'ivector-plda-scoring' HOT 12
- FileNotFoundError: [Errno 2] No such file or directory: 'ivector-plda-scoring' HOT 3
- ERRO[0001] error waiting for container: context canceled HOT 3
- Add enrollment error HOT 3
- something wrong when calculate the snr
- The data in data.tgz/test-set/61 were wrong HOT 8
- Issue at step 8 test.py HOT 10
- problem in step 8 running 'test.py' about 'data' HOT 4
- test.py: No such file or directory: 'ivector-plda-scoring' HOT 3
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from fakebob.