Is there a convenient way to switch, and/or activate and deactivate lanugage models during continuous rec runtime?
What I'm trying to do is implement load and unload model functions for an interface.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2004-06-08
Before you can expect an answer, you should tell us which recognizer you wish to use -- Sphinx2, 3, 4. (If the answer is Sphinx2, the answer is yes; I don't know about the others.)
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I'm using sphinx2, where should I be looking to be able to change language models at runtime?
fbs_init does the whole job of turning on the engine, and I can change the params passed to fbs_init, after using fbs_end(), but is there a faster/more elegant way to do this?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Anonymous
-
2004-06-15
See the Sphinx2 User Guide http://www.speech.cs.cmu.edu/sphinx/doc/sphinx2.html at the end of the section, "Speech-To_text Decoding,", beginning with "Multiple, named LMs can be...". There are four API calls that let you do just what you want.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Is there a convenient way to switch, and/or activate and deactivate lanugage models during continuous rec runtime?
What I'm trying to do is implement load and unload model functions for an interface.
Before you can expect an answer, you should tell us which recognizer you wish to use -- Sphinx2, 3, 4. (If the answer is Sphinx2, the answer is yes; I don't know about the others.)
I'm using sphinx2, where should I be looking to be able to change language models at runtime?
fbs_init does the whole job of turning on the engine, and I can change the params passed to fbs_init, after using fbs_end(), but is there a faster/more elegant way to do this?
See the Sphinx2 User Guide http://www.speech.cs.cmu.edu/sphinx/doc/sphinx2.html at the end of the section, "Speech-To_text Decoding,", beginning with "Multiple, named LMs can be...". There are four API calls that let you do just what you want.