Live Testing with Julian

1. Update your julian.jconf file (and make sure you Julian grammar files are in the adapt directory).

2. Create 'sample' grammar files similar to the ones from your How-to or Tutorial - they should look something like this:

Note: these sample grammar files use different pronunciations than your How-to and Tutorial pronunciations.  The VoxForge Speaker Independent Acoutic Model uses the larger CMU dictionnary, whereas the How-to and Tutorial uses the smaller Switchboard dictionnary, which has slightly different pronunciations than CMU's dictionnary.

sample.grammar:

S : NS_B SENT NS_E SENT: CALL_V NAME_N SENT: DIAL_V DIGIT

sample.voca:

% NS_B <s>        sil % NS_E </s>        sil % CALL_V PHONE        f ow n CALL        k ao l % DIAL_V DIAL        d ay ax l % NAME_N STEVE        s t iy v YOUNG        y ah ng % DIGIT FIVE        f ay v FOUR       f ao r NINE        n ay n EIGHT        ey t OH        ow ONE        w ah n SEVEN        s eh v ih n SIX        s ih k s THREE        th r iy TWO        t uw ZERO(2)    z ih r ow

3. Compile your grammar files (sample.grammar and sample.voca) with the mkdfa.pl script as follows (note: using the HTK 3.3 version of this script is OK):

$mkdfa.pl sample sample.grammar has 3 rules sample.voca    has 6 categories and 18 words --- Now parsing grammar file Now modifying grammar to minimize states[-1] Now parsing vocabulary file Now making nondeterministic finite automaton[6/6] Now making deterministic finite automaton[6/6] Now making triplet list[6/6] --- generated: sample.dfa sample.term sample.dict

 

Make sure your Microphone volume is similar to when you created your audio files. Then run Julius with:

$julius-4.3.1 -input mic -C sample.jconf