From documentation i understand that to begin search(decoding) from language model we have to pass searching term first, like for keyword we say "ok mighty computer" and it responds, and if we want to search from language model we have to do this recognizer.addNgramSearch(FORECAST_SEARCH, languageModel);there must be send something first in order to initialize(I'm confused, m i right or wrong, please correct me),
I want do speech recognition as soon as activity starts, whatever we say it should print on the screen when stop speaking, how can i do this
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I understand that I need a language model and this is what we have en-us.lm.bin but I dnt want but what is NGRAM SEARCH, what it holds?
// Create keyword-activation search.
recognizer.addNgramSearch(NGRAM_SEARCH, new File(assetsDir, "en-us.lm.bin"););
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
From documentation i understand that to begin search(decoding) from language model we have to pass searching term first, like for keyword we say "ok mighty computer" and it responds, and if we want to search from language model we have to do this
recognizer.addNgramSearch(FORECAST_SEARCH, languageModel);there must be send something first in order to initialize(I'm confused, m i right or wrong, please correct me),I want do speech recognition as soon as activity starts, whatever we say it should print on the screen when stop speaking, how can i do this
http://stackoverflow.com/questions/25949295/cmusphinx-pocketsphinx-recognize-all-or-large-amount-of-words
I understand that I need a language model and this is what we have en-us.lm.bin but I dnt want but what is NGRAM SEARCH, what it holds?
// Create keyword-activation search.
recognizer.addNgramSearch(NGRAM_SEARCH, new File(assetsDir, "en-us.lm.bin"););
http://cmusphinx.sourceforge.net/wiki/tutorialpocketsphinx#searches
So it means its just a string reference and we can start listening with this reference disrectly?
where its available "en-us.lm.bin"
I'm getting this error in do in background " RuntimeException: new_Decoder returned -1 , PocketSphinxJNI.new_Decoder__SWIG_1"
You can find additional details in logcat output.
I corrected and now its listening, but accurcy is so poor
@sarah how did you solved that error?