Hi Everybody,
I have trained successfully my own database in italian language using
SphinxTrainer downloaded from this page http://www.speech.cs.cmu.edu/sphinx/t
utorial.html#SphinxTrain
I have also installed Sphinx3 3.7 following the Robust Group tutorial
Anyway when i give the command ./scripts_pl/decode/slave.pl i have an error
that i post
MODULE: DECODE Decoding using models previously trained
Can't open perl script
"/home/marco/SpeechRecognition/evalita/scripts_pl/decode/s3decode.pl": Nessun
file o directory Aligning results to find error rate
Can't open file
/home/marco/SpeechRecognition/evalita/sphinx3/result/evalita.match
I have not folder result and so this .match file and i don't understand why
I have also installed tool sclite and i have changed sphinx_decode.cfg file
inserting the following string
$DEC_CFG_ALIGN = "/home/marco/sctk-2.4.0/bin/sclite";
and commenting on the string #$DEC_CFG_ALIGN = "builtin";
Thanks to everybody for any suggestion
Marco
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
thanks for the suggestion
anyway during training phase i followed also the tutorial you suggested http:
//cmusphinx.sourceforge.net/wiki/tutorialam
my problem is that i have not the folder result and the file.match and i don't
understand the reason
I followed the tutorial for database an4 and decoding worked, but for my
database not
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
i followed your suggestion so i installed SphinxTrain1.0, sphinxbase-0.6.1. As
i want use sphinx3 i installed sphinx 3.5.
I followed the new tutorial [http://cmusphinx.sourceforge.net/wiki/tutorialam]
(http://cmusphinx.sourceforge.net/wiki/tutorialam) but at the end i have
always the same problem
marco@marco-laptop:~/SpeechRecognition/evalita$ ./scripts_pl/decode/slave.pl
MODULE: DECODE Decoding using models previously trained
Aligning results to find error rate
Can't open file /.match
so i can't decode even if i have trained successfully
i wanted ask you what's the most stable version for sphinx3
With Regards
Marco
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
but at the end i have always the same problem marco@marco-
laptop:~/SpeechRecognition/evalita$ ./scripts_pl/decode/slave.pl MODULE:
DECODE Decoding using models previously trained Aligning results to find error
rate Can't open file /.match
Looks like a configuration problem. Most likely it's caused by obsolete
sphinx3 version. It's sad you create your problems yourself.
wanted ask you what's the most stable version for sphinx3
Hi Everybody,
I have trained successfully my own database in italian language using
SphinxTrainer downloaded from this page http://www.speech.cs.cmu.edu/sphinx/t
utorial.html#SphinxTrain
I have also installed Sphinx3 3.7 following the Robust Group tutorial
Anyway when i give the command ./scripts_pl/decode/slave.pl i have an error
that i post
MODULE: DECODE Decoding using models previously trained
Can't open perl script
"/home/marco/SpeechRecognition/evalita/scripts_pl/decode/s3decode.pl": Nessun
file o directory Aligning results to find error rate
Can't open file
/home/marco/SpeechRecognition/evalita/sphinx3/result/evalita.match
I have not folder result and so this .match file and i don't understand why
I have also installed tool sclite and i have changed sphinx_decode.cfg file
inserting the following string
$DEC_CFG_ALIGN = "/home/marco/sctk-2.4.0/bin/sclite";
and commenting on the string #$DEC_CFG_ALIGN = "builtin";
Thanks to everybody for any suggestion
Marco
That's outdated page, not sure why you decided to follow that. Up-to-date
tutorial is here:
http://cmusphinx.sourceforge.net/wiki/tutorialam
thanks for the suggestion
anyway during training phase i followed also the tutorial you suggested http:
//cmusphinx.sourceforge.net/wiki/tutorialam
my problem is that i have not the folder result and the file.match and i don't
understand the reason
I followed the tutorial for database an4 and decoding worked, but for my
database not
Elementary Watson
The reason you don't have result is that you don't have script s3decode.pl in
place.
The reason you don't have script is that you didn't run setup_sphinx script in
decoder.
The reason you did run the setup script is that you didn't actually follow the
tutorial.
hi nshmyrev
i followed your suggestion so i installed SphinxTrain1.0, sphinxbase-0.6.1. As
i want use sphinx3 i installed sphinx 3.5.
I followed the new tutorial [http://cmusphinx.sourceforge.net/wiki/tutorialam]
(http://cmusphinx.sourceforge.net/wiki/tutorialam) but at the end i have
always the same problem
marco@marco-laptop:~/SpeechRecognition/evalita$ ./scripts_pl/decode/slave.pl
MODULE: DECODE Decoding using models previously trained
Aligning results to find error rate
Can't open file /.match
so i can't decode even if i have trained successfully
i wanted ask you what's the most stable version for sphinx3
With Regards
Marco
Looks like a configuration problem. Most likely it's caused by obsolete
sphinx3 version. It's sad you create your problems yourself.
Latest released one is 0.8
http://sourceforge.net/projects/cmusphinx/files/sphinx3/0.8/sphinx3-0.8.tar.g
z/download