getalp / ALFFA_PUBLIC

MIT License
48 stars 98 forks source link

Cannot find lm/amharic.lm.data.segmented file #3

Open Habush opened 7 years ago

Habush commented 7 years ago

In the 03_LM.sh script found in kaldi-scripts folder there is a line that uses srilm. Here is the line: _/home/melese/toolkit/srilm/bin/i686-m64/ngram-count -order 5 -text lm/amharic.lm.data.segmented -lm lm/amharic.train.lm.data.arpa -unk -kndiscount1 -kndiscount2 -kndiscount3 -kndiscount4 -kndiscount5 -gt1min 1 -gt2min 1 -gt3min 1 -gt4min 1 -gt5min 1_

However, I cannot find the lm/amharic.lm.data.segmented file. In the lm folder there are only two zip files, namely amharic.PART1.zip and amharic.PART2.zip . I amharic.lm.data.segmented file cannot be found even if I extract the zip files.

besacier commented 7 years ago

Hi

You can just unzip the two files and concatenate them in a single one renamed amharic.lm.data.segmented

It should work then

Best

L

(Gesendet von meinem Ipad)

Le 23 nov. 2016 à 12:52, Abdulrahman Semrie notifications@github.com a écrit :

In the 03_LM.sh script found in kaldi-scripts folder there is a line that uses srilm. Here is the line: /home/melese/toolkit/srilm/bin/i686-m64/ngram-count -order 5 -text lm/amharic.lm.data.segmented -lm lm/amharic.train.lm.data.arpa -unk -kndiscount1 -kndiscount2 -kndiscount3 -kndiscount4 -kndiscount5 -gt1min 1 -gt2min 1 -gt3min 1 -gt4min 1 -gt5min 1

However, I cannot find the lm/amharic.lm.data.segmented file. In the lm folder there are only two zip files, namely amharic.PART1.zip and amharic.PART2.zip . I amharic.lm.data.segmented file cannot be found even if I extract the zip files.

— You are receiving this because you are subscribed to this thread. Reply to this email directly, view it on GitHub, or mute the thread.