Menu

A question about the sphinx3 library.

Help
chris
2007-12-25
2012-09-22
  • chris

    chris - 2007-12-25

    As you know,when I init sphinx3 library,I can use cmd_ln_str() or cmd_ln_parse_file(),some parameters is quite sample just some flag or value(like -feat,-op_mode..),but for the accoustic model,it just a path to a file,the library will read the file.In my opinion,the library should not use the file directly, right?When I wrote my application,infact the accoustic model is in the memory(also the fsg grammar and dictionary file),I have to use a temp file to save it to the disc,and use them.is there a easy way to use the memory model?maybe I can build a link list or something else data structure?

     
    • chris

      chris - 2007-12-28

      Oh,Thanks.
      but How about use memory accoustic modle?is there any easy way?
      I cheched part of the source code,I think I should modify the parse part in many files,I really do't what to do that:(
      but it seems I have to.

       
    • chris

      chris - 2007-12-25

      And by the way,is there any API reference for sphinx3?

       
      • David Huggins-Daines

        Actually, we do have Doxygen generated documentation on-line:

        http://www.speech.cs.cmu.edu/sphinx/sphinx3/doxygen/html/

        It is a bit difficult to navigate though.

         
      • Nickolay V. Shmyrev

        No, sorry. It would be nice to create one with say doxygen.

         

Log in to post a comment.