android - CMUSphinx PocketSphinx - Recognize all (or large amount) of words -

before tried used pocketsphinx android, used google's voice recognition api. didn't need set search name or dictionary file. recognized every word told.

now, in pocketsphinx, need it. can find how set recognition 1 word, or set dictionary (the ones available in demo project have few words) recognizer think these words exist, means if says similar, recognizer thinks word listed in dictionary.

i want ask, how set few search names, or how set recognize words available (or large amount of them)? maybe has dictionary file big number of words?

before tried used pocketsphinx android, used google's voice recognition api. didn't need set search name or dictionary file. recognized every word told.

google api recognizes large still limited set of words too. long time failed recognize "spotify". google offline speech recognizer uses 50k words described in publication.

i want ask, how set few search names, or how set recognize words available (or large amount of them)? maybe has dictionary file big number of words?

demo includes large vocabulary speech recognition language model (forecast part). there bigger language model english language available download, example en-us generic language model.

the simple code run recognition that:

 recognizer = defaultsetup()    .setacousticmodel(new file(assetsdir, "en-us-ptm"))    .setdictionary(new file(assetsdir, "cmudict-en-us.dict"))    .getrecognizer();   recognizer.addlistener(this);    // create keyword-activation search.   recognizer.addngramsearch(ngram_search, new file(assetsdir, "en-us.lm.bin"););    // start search   recognizer.startlistening(ngram_search); 

however, not easy fit device , decode in realtime. if want decode speech in realtime large vocabulary need stream audio server. or need restrict vocabulary , language small subset of generic english. can learn more speech recognition in cmusphinx in tutorial.