Last month I have finally build my own acoustic model about chess pieces and
it work really good.
However, now I start to build the new model about walk, left, right something
like that.
so I try to use my previous model and do this step.
but it also gives the same error.
anyway, I ignore that because it a step that check.
after that check it with pocketsphinx_continuous with those new model
but every word I said, It gives no answer.
I think I did the same thing as my previous work.
Anyway how to fix this?
Thank you.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have manage to solve that problem.
The problem that it gives no result is because my language model file that I
generate from cmuclmtk.
But instead of use cmuclmtk to generate language model file, I have used
online language model tool from this link http://www.speech.cs.cmu.edu/tools/lmtool-
new.html
and it gives the acceptable accuracy.
Do something wrong occur in my cmuclmtk file?
Thank you
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have generated the language model from http://www.speech.cs.cmu.edu/tools
/lmtool-new.html itself
but still I am getting the exact same error. Do you have an idea of some other
reason that might me causing this error?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I have generated the language model from http://www.speech.cs.cmu.edu/tools
/lmtool-new.html itself
but still I am getting the exact same error. Do you have an idea of some other
reason that might me causing this error?
Sorry, until you provide more information (at least pocketsphinx decoder log
from the logdir/decode subfolder) it's hard to say something. Information is
critical to solve th eproblem, you should know that.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hello, I'm here again.
Last month I have finally build my own acoustic model about chess pieces and
it work really good.
However, now I start to build the new model about walk, left, right something
like that.
It passed
with no problem occur.
then I do
I have got an error look like this
so I try to use my previous model and do this step.
but it also gives the same error.
anyway, I ignore that because it a step that check.
after that check it with pocketsphinx_continuous with those new model
but every word I said, It gives no answer.
I think I did the same thing as my previous work.
Anyway how to fix this?
Thank you.
Thx Mr.nshmyrev for the answer
I have manage to solve that problem.
The problem that it gives no result is because my language model file that I
generate from cmuclmtk.
But instead of use cmuclmtk to generate language model file, I have used
online language model tool from this link
http://www.speech.cs.cmu.edu/tools/lmtool-
new.html
and it gives the acceptable accuracy.
Do something wrong occur in my cmuclmtk file?
Thank you
Who knows, it depends on what exact commands you are running.
I have generated the language model from http://www.speech.cs.cmu.edu/tools
/lmtool-new.html itself
but still I am getting the exact same error. Do you have an idea of some other
reason that might me causing this error?
Sorry, until you provide more information (at least pocketsphinx decoder log
from the logdir/decode subfolder) it's hard to say something. Information is
critical to solve th eproblem, you should know that.