2

Though I was having some problems to make sphinx 4 work initially but now its working great. Since the grammar is finite whether it is JSGF or N-Gram, I am not able to make sphinx recognize every possible word or sentence. I want to build something like voice based google search. Since the search may have any possible word or combination of words. It is troublesome to add all the dictionary words in grammar. Kindly help.

barryhunter
  • 20,886
  • 3
  • 30
  • 43
utsav
  • 71
  • 1
  • 7

1 Answers1

3

No, it's not possible. Every speech recognizer including the one from Google uses limited vocabulary. Google uses large one of about 1.5 million words but still limited. So CMUSphinx. You can verify that by trying to recognize rare proper names, it's almost impossible even with Google.

The practical solution is to use large language model with large vocabulary.

It's a open research question to detect new words in an audio stream and add them recognizer with spoken or other type of feedback.

Nikolay Shmyrev
  • 24,897
  • 5
  • 43
  • 87
  • Thankyou @Nikolay. Is there any limitation on number of words that can be added to grammar? I am trying to add about thousand of words in grammar but on executing it is giving Exception. It is working fine with 30 or 40 words. Is it possible to add all the words in dictionary to grammar? I am adding words in the following way: public = ((words added here separated with |)*); – utsav Aug 14 '13 at 17:28
  • There is no limitation, you can add as many words as you want. – Nikolay Shmyrev Aug 15 '13 at 06:49