Julius Startup Output

C:>julius-4.3.1 -input mic -C Sample.jconf
STAT: include config: Sample.jconf
STAT: jconf successfully finalized
STAT: *** loading AM00 _default
Stat: init_phmm: Reading in HMM definition
Stat: rdhmmdef: ascii format HMM definition
Stat: rdhmmdef: limit check passed
Stat: check_hmm_restriction: an HMM with several arcs from initial state found:
Stat: rdhmmdef: this HMM requires multipath handling at decoding
Stat: rdhmmdef: no <SID> embedded
Stat: rdhmmdef: assign SID by the order of appearance
Stat: init_phmm: defined HMMs:    46
Stat: init_phmm: loading ascii hmmlist
Stat: init_phmm: logical names:   489 in HMMList
Stat: init_phmm: base phones:    41 used in logical
Stat: init_phmm: finished reading HMM definitions
STAT: m_fusion: force multipath HMM handling by user request
STAT: making pseudo bi/mono-phone for IW-triphone
Stat: hmm_lookup: 364 pseudo phones are added to logical HMM list
STAT: *** AM00 _default loaded
STAT: *** loading LM00 _default
STAT: reading [sample.dfa] and [sample.dict]...
Stat: init_voca: read 18 words
STAT: done
STAT: Gram #0 sample registered
STAT: Gram #0 sample: new grammar loaded, now mash it up for recognition
STAT: Gram #0 sample: extracting category-pair constraint for the 1st pass
STAT: Gram #0 sample: installed
STAT: Gram #0 sample: turn on active
STAT: grammar update completed
STAT: *** LM00 _default loaded
STAT: ------
STAT: All models are ready, go for final fusion
STAT: [1] create MFCC extraction instance(s)
STAT: *** create MFCC calculation modules from AM
STAT: AM 0 _default: create a new module MFCC01
STAT: 1 MFCC modules created
STAT: [2] create recognition processing instance(s) with AM and LM
STAT: composing recognizer instance SR00 _default (AM00 _default, LM00 _default)

STAT: Building HMM lexicon tree
STAT: lexicon size: 210 nodes
STAT: coordination check passed
STAT: multi-gram: beam width set to 200 (guess) by lexicon change
STAT: wchmm (re)build completed
STAT: SR00 _default composed
STAT: [3] initialize for acoustic HMM calculation
Stat: outprob_init: state-level mixture PDFs, use calc_mix()
Stat: addlog: generating addlog table (size = 1953 kB)
Stat: addlog: addlog table generated
STAT: [4] prepare MFCC storage(s)
STAT: [5] prepare for real-time decoding
STAT: All init successfully done

STAT: ###### initialize input device
----------------------- System Information begin ---------------------
JuliusLib rev.4.3.1 (fast)

Engine specification:
 -  Base setup   : fast
 -  Supported LM : DFA, N-gram, Word
 -  Extension    : NoPThread
 -  Compiled by  : i686-w64-mingw32-gcc -O6 -fomit-frame-pointer

Configuration of Modules

 Number of defined modules: AM=1, LM=1, SR=1

 Acoustic Model (with input parameter spec.):
 - AM00 "_default"

 Language Model:
 - LM00 "_default"
        grammar #1:
            dfa  = sample.dfa
            dict = sample.dict

 - SR00 "_default" (AM00, LM00)

Speech Analysis Module(s)

[MFCC01]  for [AM00 _default]

 Acoustic analysis condition:
               parameter = MFCC_0_D_N_Z (25 dim. from 12 cepstrum + c0, abs ener
gy supressed with CMN)
        sample frequency = 16000 Hz
           sample period =  625  (1 = 100ns)
             window size =  400 samples (25.0 ms)
             frame shift =  160 samples (10.0 ms)
            pre-emphasis = 0.97
            # filterbank = 24
           cepst. lifter = 22
              raw energy = False
        energy normalize = False
            delta window = 2 frames (20.0 ms) around
             hi freq cut = OFF
             lo freq cut = OFF
         zero mean frame = OFF
               use power = OFF
                     CVN = OFF
                    VTLN = OFF

    spectral subtraction = off

 cep. mean normalization = yes, real-time MAP-CMN, updating mean with last 0.0 s
ec. input
  initial mean from file = N/A
   beginning data weight = 100.00
 cep. var. normalization = no

         base setup from = Julius defaults

Acoustic Model(s)

[AM00 "_default"]

 HMM Info:
    46 models, 126 states, 126 mpdfs, 126 Gaussians are defined
              model type = context dependency handling ON
      training parameter = MFCC_N_D_Z_0
           vector length = 25
        number of stream = 1
             stream info = [0-24]
        cov. matrix type = DIAGC
           duration type = NULLD
        max mixture size = 1 Gaussians
     max length of model = 5 states
     logical base phones = 41
       model skip trans. = exist, require multi-path handling
      skippable models = sp (1 model(s))

 AM Parameters:
        Gaussian pruning = safe  (-gprune)
  top N mixtures to calc = 2 / 0  (-tmix)
    short pause HMM name = "sp" specified, "sp" applied (physical)  (-sp)
  cross-word CD on pass1 = handle by approx. (use max. prob. of same LC)
   sp transition penalty = -70.0

Language Model(s)

[LM00 "_default"] type=grammar

 DFA grammar info:
      6 nodes, 6 arcs, 6 terminal(category) symbols
      category-pair matrix: 32 bytes (712 bytes allocated)

 Vocabulary Info:
        vocabulary size  = 18 words, 52 models
        average word len = 2.9 models, 8.7 states
       maximum state num = 15 nodes per word
       transparent words = not exist
       words under class = not exist

   found sp category IDs =


[SR00 "_default"]  AM00 "_default"  +  LM00 "_default"

 Lexicon tree:
         total node num =    210
          root node num =     18
          leaf node num =     18

        (-penalty1) IW penalty1 = +5.0
        (-penalty2) IW penalty2 = +20.0
        (-cmalpha)CM alpha coef = 0.050000

         inter-word short pause = on (append "sp" for each word tail)
          sp transition penalty = -70.0
 Search parameters:
            multi-path handling = yes, multi-path mode enabled
        (-b) trellis beam width = 200 (-1 or not specified - guessed)
        (-bs)score pruning thres= disabled
        (-n)search candidate num= 1
        (-s)  search stack size = 500
        (-m)    search overflow = after 2000 hypothesis poped
                2nd pass method = searching sentence, generating N-best
        (-b2)  pass2 beam width = 200
        (-lookuprange)lookup range= 5  (tm-5 <= t <tm+5)
        (-sb)2nd scan beamthres = 200.0 (in logscore)
        (-n)        search till = 1 candidates found
        (-output)    and output = 1 candidates out of above
         IWCD handling:
           1st pass: approximation (use max. prob. of same LC)
           2nd pass: loose (apply when hypo. is popped and scanned)
         all possible words will be expanded in 2nd pass
         build_wchmm2() used
         lcdset limited by word-pair constraint
        short pause segmentation = off
        fall back on search fail = off, returns search failure

Decoding algorithm:

        1st pass input processing = real time, on-the-fly
        1st pass method = 1-best approx. generating indexed trellis
        output word confidence measure based on search-time scores


 Input stream:
                     input type = waveform
                   input source = microphone
            device API          = default
                  sampling freq. = 16000 Hz
                 threaded A/D-in = not supported (live input may be dropped)
           zero frames stripping = on
                 silence cutting = on
                     level thres = 2000 / 32767
                 zerocross thres = 60 / sec.
                     head margin = 300 msec.
                     tail margin = 400 msec.
                      chunk size = 1000 samples
            long-term DC removal = off
            long-term DC removal = off
            level scaling factor = 1.00 (disabled)
              reject short input = off
              reject  long input = off

----------------------- System Information end -----------------------

Notice for feature extraction (01),
        * Cepstral mean normalization for real-time decoding:       *
        * NOTICE: The first input may not be recognized, since      *
        *         no initial mean is available on startup.          *

### read waveform input
Stat: adin_portaudio: audio cycle buffer length = 256000 bytes
Stat: adin_portaudio: sound capture devices:
  1 [MME: Microsoft Sound Mapper - Input]
  2 [MME: Microphone (USB Audio Device)]
  6 [Windows DirectSound: Primary Sound Capture Driver]
  7 [Windows DirectSound: Microphone (USB Audio Device)]
Stat: adin_portaudio: APIs: DirectSound MME
Stat: adin_portaudio: -- DirectSound selected
Stat: adin_portaudio: [Windows DirectSound: Primary Sound Capture Driver]
Stat: adin_portaudio: (you can specify device by "PORTAUDIO_DEV_NUM=number"
Stat: adin_portaudio: try to set default low latency from portaudio: 0 msec
Stat: adin_portaudio: latency was set to 0.000000 msec
<<< please speak >>>