Please show me how to get started (like a link to good instructions) with cmusphinx or else recommend a different voice recognition program that works.
thanks.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Answer given by Nickolay that resolved it:
You need to replace $SPHINX_ROOT with the real folder in a cfg file yourself. Of course sphinx doesn't do any shell substitutions.
You could get this information just by reading your log:
ERROR: "cmd_ln.c", line 718: Cannot open configuration file $SPHINX_ROOT/sphinx3/model/hmm/hub4_cd_continuous_8gau_1s_c_d_dd/feat.params for reading
v
ALso, you say you want to use it for personal use as a transcriber tool i assume, and not as a developer tool. You might just want to consider installing dragon naturally speaking via wine.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
thanks, but a dragon/windows combination is very expensive, requires lot of cpu, isn't customizable and still sucks. I have used dragon extensively at work. It integrates so tightly with microsof os, I just can't believe it would work under wine, and the wine database says it doesn't work.
why not just get sphinx working and support open source? I want a lightweight transcriber that can replace my keyboard, not a huge app that tries to be everything to every window and my os.
My goal is to get gnu/linux transcriber working inside the bash shell and teach it 500 common words. From inside the bash shell I can do email, maintain my websites, and access the internet.
After my transcriber works in the bash shell the next step is to setup my transcriber on a mini laptop that I can pack with me.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I need to make cmusphinx or other voice recognition work on my computer. I really need it because my hands hurt from typing so much.
I am using xubuntu right now but i am flexible if another setup works better.
I already tried installing cmusphinx 3. I followed the instructions at this url:
http://sphinx.subwiki.com/sphinx/index.php/Hello_World_Decoder_QuickStart_Guide
I got this error, and I didn't understand it because I did
download the def file, make, and convert to binary dmp as instructed.
"FATAL_ERROR: "mdef.c", line 679: No mdef-file"
I got so frustrated I gave up and uninstalled it. Then I found this forum.
Can anyone please help me get started?
What kind of help do you need? The advise to return and try once again and read the text carefully? Or the link on the howto on asking questions?
Hi, I basically need newbie help for CMUsphinx or another voice recognition program for gnu/linux that works.
I already followed the instructions at http://sphinx.subwiki.com/sphinx/index.php/Hello_World_Decoder_QuickStart_Guide twice to the letter and received the error that I put in my first post. I don't see how going through those steps would help.
Please show me how to get started (like a link to good instructions) with cmusphinx or else recommend a different voice recognition program that works.
thanks.
Searching the forum I found the exact same question asked before:
https://sourceforge.net/forum/message.php?msg_id=5960287
Answer given by Nickolay that resolved it:
You need to replace $SPHINX_ROOT with the real folder in a cfg file yourself. Of course sphinx doesn't do any shell substitutions.
You could get this information just by reading your log:
ERROR: "cmd_ln.c", line 718: Cannot open configuration file $SPHINX_ROOT/sphinx3/model/hmm/hub4_cd_continuous_8gau_1s_c_d_dd/feat.params for reading
v
Or post more of your log. As for alternatives that you ask for above, you may want to try sphinx4. It is easier to get setup imo. http://www.speech.cs.cmu.edu/sphinx/tutorial.html
Also sphinx4 has the dictator project that is the closest thing to a transcriber.
http://www.speech.cs.cmu.edu/sphinx/dictator/
ALso, you say you want to use it for personal use as a transcriber tool i assume, and not as a developer tool. You might just want to consider installing dragon naturally speaking via wine.
thanks, but a dragon/windows combination is very expensive, requires lot of cpu, isn't customizable and still sucks. I have used dragon extensively at work. It integrates so tightly with microsof os, I just can't believe it would work under wine, and the wine database says it doesn't work.
why not just get sphinx working and support open source? I want a lightweight transcriber that can replace my keyboard, not a huge app that tries to be everything to every window and my os.
My goal is to get gnu/linux transcriber working inside the bash shell and teach it 500 common words. From inside the bash shell I can do email, maintain my websites, and access the internet.
After my transcriber works in the bash shell the next step is to setup my transcriber on a mini laptop that I can pack with me.
Well, you are welcome to implement such dictation software. Take a look on voicekeyboard btw:
https://sourceforge.net/project/showfiles.php?group_id=227605