Closed lyxminnie closed 7 years ago
Hi @lyxminnie,
It seems like you're installing software from Homebrew. The error messages suggest that the SphinxBase version in use is quite old, since it doesn't support that newer format.
For SphinxBase, Homebrew has version 0.8, which is from 2012. This is why the docs outline how to build SphinxBase from source. If you follow those instructions instead, it should be possible to build the model.
Please re-open this issue if the errors persist! I am planning to eventually provide a Docker image with all of these tools to simplify model building.
I tried to install from the source code, and it works! Thank you!
Hi,
I am getting the same error as well. I have built sphinxbase from source code as given the docs. My system is also Mac OS . INFO: ngram_model_arpa.c(292): Reading trigrams INFO: ngram_model_arpa.c(555): 21488 = #trigrams created INFO: ngram_model_arpa.c(556): 144 = #prob3 entries ERROR: "ngram_model.c", line 183: language model file type not supported ERROR: "sphinx_lm_convert.c", line 212: Failed to write language model in format (null) to hindi.lm.bin
which sphinx_lm_convert /usr/local/bin/sphinx_lm_convert
Please advise.
Hi,
I was able to get this working. Had to remove --force from the command to build and make.
Regards, Ashok
Steps to reproduce
Expected behaviour
step 1-3 succeeded. step 4 should generate chinese.lm.bin.
Actual behaviour
sphinx_lm_convert -i chinese.lm -o chinese.lm.bin INFO: cmd_ln.c(691): Parsing command line: sphinx_lm_convert \ -i chinese.lm \ -o chinese.lm.bin
Current configuration: [NAME] [DEFLT] [VALUE] -case -debug 0 -help no no -i chinese.lm -ienc -ifmt -logbase 1.0001 1.000100e+00 -mmap no no -o chinese.lm.bin -oenc utf8 utf8 -ofmt
INFO: ngram_model_arpa.c(477): ngrams 1=63944, 2=16600781, 3=20708460 INFO: ngram_model_arpa.c(135): Reading unigrams INFO: ngram_model_arpa.c(516): 63944 = #unigrams created INFO: ngram_model_arpa.c(195): Reading bigrams .............................................................................................................................................................................................................................................................INFO: ngram_model_arpa.c(533): 16600781 = #bigrams created INFO: ngram_model_arpa.c(534): 32337 = #prob2 entries INFO: ngram_model_arpa.c(542): 24468 = #bo_wt2 entries INFO: ngram_model_arpa.c(292): Reading trigrams ...........................................................................................................................................................................................................................................................................................................................INFO: ngram_model_arpa.c(555): 20708460 = #trigrams created INFO: ngram_model_arpa.c(556): 27937 = #prob3 entries ERROR: "ngram_model.c", line 183: language model file type not supported ERROR: "sphinx_lm_convert.c", line 212: Failed to write language model in format (null) to chinese.lm.bin
System information
(Delete all the statements that don't apply.)
My system is <Mac OS X EI Capitan 10.11.6>. (For example, "Ubuntu 16.04 LTS x64", "Windows 10 x64", or "macOS Sierra".)
My Python version is <Python 3.5.2 :: Anaconda custom (x86_64)>. (You can check this by running
python -V
.)My Pip version is <pip 8.1.1 from /anaconda/lib/python3.5/site-packages (python 3.5)>. (You can check this by running
pip -V
.)My SpeechRecognition library version is <3.6.5>. (You can check this by running
python -c "import speech_recognition as sr;print(sr.__version__)"
.)I installed PocketSphinx from. (For example, from the Debian repositories, from Homebrew, or from the source code.)