====== How To Use ======
-
Put the whole directory under /var/www/ (apache-directory) or using a symbolic link (ln -s source dest)
-
Install dependencies a) bin: sox (Sound eXchange, see http://sox.sourceforge.net/) b) python: networkx, numpy (sudo easy_install networkx, sudo easy_install numpy) c) node.js: express (npm install express)
-
Run the server by $ node node.js/server.js (Of course you can put it in screen to run in the background)
-
Open index.html
====== Files & Directories ====== FOLDER DESCRIPTIONS useless/ some intermediate files under development, not sure it's ok to delete them or not user_wav/ TEMP directory. Wav recored during DEMO lies in here. It's ok to remove them when the server is not on, and therefore ignored by .gitignore. user_score/ TEMP directory. Scores analyzed by myct_graders. Kind of the same as above.
==== DEMO client ==== index.html js/ css/ Cascade Style Sheet (CSS) img/ common/ Common JavaScript libraries used by either *.html or *.js
==== DEMO server ==== go.py Python scripts. node.js/ |_ server.js Main server program connecting the client with the core programs node_modules/ Node modules. Eg. express (You can also install the latest one using cmd "npm install express", but there's no guarantee of stability due to discrepencies among different versions.)
==== Core Programs ==== myct_grader/ provided by MyET. (No source codes available) |_ myct_grader A binary executable used for grading the user's Chinese based on 1) Pronunciation 2) Tone 3) Timing 4) Emphasis (compiling environment: Linux 64bit executable Linux speech-ubuntu 3.0.0-12-generic #20-Ubuntu SMP Fri Oct 7 14:56:25 UTC 2011 x86_64 x86_64 x86_64 GNU/Linux) ASAS-cdb.hmm HMM needed by binary myct_grader LCP_Dictory.txt Another core file needed by binary myct_grader
algorithms/ Core algorithms. dialogueTreeFile/ DialogueTree and core data. dialogueTree_owv/ dialogueTree_wav/
==== Others ==== less/ .gitignore