I want to use the zh_broadcastnews_64000_utf8.DMP language model in pocketsphinx, but I recognized the result could not be normally presented.I'm using it on windows .Should I convert the utf8 format to GBK foramt?
If true, should I use the sphinx_lm_convert tool to change it to lm file and convert it? but when I try to convert,it would get stuck in the progress meanwhile the CPU keeping busy and the lm file doesn't increase any more.
Beg for your help sincerely,thanks!
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Thanks! I did it in sphinx_lm_convert the release version before .Now I have tried it in debug version and it works successfully now. Thanks for your help !
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I want to use the zh_broadcastnews_64000_utf8.DMP language model in pocketsphinx, but I recognized the result could not be normally presented.I'm using it on windows .Should I convert the utf8 format to GBK foramt?
If true, should I use the sphinx_lm_convert tool to change it to lm file and convert it? but when I try to convert,it would get stuck in the progress meanwhile the CPU keeping busy and the lm file doesn't increase any more.
Beg for your help sincerely,thanks!
Yes
No, you can just convert the output
How to convert the DMP file directly?
Could you give me more details? It look like I cannot convert the encode of DMP file directly in programs like notepad++
To convert DMP to ARPA, run
sphinx_lm_convert -i /path/to/DMP -ifmt DMP -o /path/to/LM -ofmt ARPA
But using the command you given there will be a phenomenon like I have mentioned above ..the ARPA file cannot be generated completely
I just now did the same and an ARPA file was successfully generated. It took some time, but the file was created. Get more RAM maybe.
Thanks! I did it in sphinx_lm_convert the release version before .Now I have tried it in debug version and it works successfully now. Thanks for your help !