how to use word_align.pl ?
I'am running
./word_align.pl refrence_file hypothesis_file
Also tried
./word_align.pl calculate_wer refrence_file hypothesis_file
None of the above worked . Can anyone help me the correct usage of word_align.pl
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
This align script is a bit choosy. You have bunch of empty lines at the end of adaptation-test.transcription. Remove them and it will be just fine. I got this on your files:
TOTAL Words: 33 Correct: 30 Errors: 4
TOTAL Percent correct = 90.91% Error = 12.12% Accuracy = 87.88%
TOTAL Insertions: 1 Deletions: 1 Substitutions: 2
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
how to use word_align.pl ?
I'am running
./word_align.pl refrence_file hypothesis_file
Also tried
./word_align.pl calculate_wer refrence_file hypothesis_file
None of the above worked . Can anyone help me the correct usage of word_align.pl
"refrence_file" - is a transcription file. Like in tutorial: http://www.speech.cs.cmu.edu/cmusphinx/moindocs/arctic20.transcription
"hypothesis_file" - is a file with recognition results one per line. It is generated by pocketsphinx_batch, by specifying "-hyp path/to/file"
Having this two text files you can calculate WER
I'am not confused about the files to be mentioned.
I want to know the usage of word_align.pl
The following doesnt work
./word_align.pl adaptation-test.transcription adapation-test.hyp
Please share files you trying. All three of them. And the output of the command you have on your side.
Attached are the files .
This align script is a bit choosy. You have bunch of empty lines at the end of adaptation-test.transcription. Remove them and it will be just fine. I got this on your files:
kinly plz past the exact script you run in your cmd for word_align.pl I'm also facing problem in decoding
thanks,
oh now i get it .....Thanks for your help