srvk / eesen

The official repository of the Eesen project
http://arxiv.org/abs/1507.08240
Apache License 2.0
824 stars 342 forks source link

Missing label.counts #203

Closed tony-stark1 closed 5 years ago

tony-stark1 commented 5 years ago

Hi,

I'm trying to run the WSJ example it works up until it reaches this line of code: gunzip -c $dir/labels.tr.gz | awk '{line=$0; gsub(" "," 0 ",line); print line " 0";}' | \ analyze-counts --verbose=1 --binary=false ark:- $dir/label.counts >& $dir/log/compute_label_counts.log || exit 1

At that point it fails and according to the logs it is because it cannot find the label.counts file. I looked for the file and it doesn't seem to exist. I've tried looking through the code but I can't find where this file is created.

greping the directory only show places where the file is used but not how to create it. Please let me know what to run to make this file or how to make it manually if that is the solution.

riebling commented 5 years ago

I think that file comes from here: https://github.com/srvk/eesen/blob/4a98525fabf491c32317326fdca11f44ea9117f1/asr_egs/wsj/steps/train_ctc_parallel.sh#L114-L121

tony-stark1 commented 5 years ago

So it looks like its failing to created the file due to this error: ERROR (analyze-counts:ReadBasicType():base/io-funcs-inl.h:85) Read failure in ReadBasicType, file position is -1, next char is -1 WARNING (analyze-counts:Read():util/kaldi-holder-inl.h:255) BasicVectorHolder::Read, could not interpret line: 0 40 0 40 0 40 0 20 0 F 0 40 0 40 0

Which looks like something is going wrong when creating the labels.tr.gz file

icestoneking commented 4 years ago

So it looks like its failing to created the file due to this error: ERROR (analyze-counts:ReadBasicType():base/io-funcs-inl.h:85) Read failure in ReadBasicType, file position is -1, next char is -1 WARNING (analyze-counts:Read():util/kaldi-holder-inl.h:255) BasicVectorHolder::Read, could not interpret line: 0 40 0 40 0 40 0 20 0 F 0 40 0 40 0

Which looks like something is going wrong when creating the labels.tr.gz file

hi,Is this problem solved?I also have this problem.