| $julian -input mic -C julian.jconf include config: julian.jconf ###### check configurations ###### initialize input device fragment size = 1024 bytes (10 msec) AD-in thread created ###### build up system Reading in HMM definition...(ascii)... limit check passed defined HMMs: 50 logical names: 506 in HMMList base phones: 44 used in logical done Making pseudo bi/mono-phone for IW-triphone...369 added as logical...done reading [sample.dfa] and [sample.dict]... Reading in dictionary... 18 words...done Reading in DFA grammar...done - Gram #0: read [grammars] # 0: [active ] 18 words, 6 categories, 6 nodes (new) "sample" gram "sample" registered - Grammar update check Mapping dict item <-> DFA terminal (category)...done - Gram #0: installed Building HMM lexicon tree.................153+0=153 nodes coordination check passed done now beam width = 63 (guess) - update completed [grammars] # 0: [active ] 18 words, 6 categories, 6 nodes "sample" Grobal: 18 words, 6 categories, 5 nodes Generating addlog table...1953 kb...done All init successfully done ------------- System Info begin ------------- Julian rev.3.5.2-multipath (standard) Engine configuration: - Base setup : standard - Tunings : DFA, StrictIWCD2, IconvOutput - Compiled by: gcc -O6 -fomit-frame-pointer Continuous Speech Recognition Parser based on automaton grammar Files: hmmfilename=hmm15/hmmdefs hmmmapfilename=tiedlist grammar #1: dfa = sample.dfa dict = sample.dict Acoustic analysis condition: sample period = 208 ns (48000 Hz) window size = 400 samples (8.3 ms) frame shift = 160 samples (3.3 ms) delta window = 2 frames (6.7 ms) around hi freq. = OFF lo freq. = OFF energy measure = (_0) 0'th cepstral spectral subtraction = off HMM Info: 50 models, 135 states, 135 mixtures are defined model type = context dependency handling ON training parameter = MFCC_N_D_Z_0 vector length = 25 cov. matrix type = DIAGC duration type = NULLD mixture num = 1 max state num = 5 Dictionary Info: vocabulary size = 18 words, 51 models average word len = 2.8 models, 8.5 states maximum state num = 15 nodes per word transparent words = not exist words under class = not exist Lexicon tree info: total node num = 153 root node num = 18 leaf node num = 18 DFA grammar info: 5 nodes, 6 arcs, 6 terminal(category) symbols category-pair matrix size is 4 bytes Weights and words: (-penalty1) IW penalty1 = +0.0 (-penalty2) IW penalty2 = +0.0 (-cmalpha)CM alpha coef = 0.050000 (-sp)shortpause HMM name= "sp" specified, "sp" applied (physical) found sp category IDs = Search parameters: 1st pass decoding = on-the-fly with last CMN 1st pass method = 1-best approx. generating indexed trellis (-b) trellis beam width = 63 (-1 or not specified - guessed) (-n)search candidate num= 10 (-s) search stack size = 500 (-m) search overflow = after 2000 hypothesis poped 2nd pass method = searching sentence, generating N-best (-b2) pass2 beam width = 100 (-lookuprange)lookup range= 5 (tm-5 <= t <tm+5) (-sb)2nd scan beamthres = 80.0 (in logscore) (-gprune)Gauss. pruning = none (full computation) (-n) search till = 10 candidates found (-output) and output = 1 candidates out of above IWCD handling: 1st pass: approximation (use average prob. of same LC) 2nd pass: strict (apply when expanding hypo. ) all possible words will be expanded in 2nd pass build_wchmm2() used lcdset limited by word-pair constraint output word confidence measure based on search-time scores System I/O configuration: speech input source = microphone sampling freq. = 48000 Hz threaded A/D-in = supported, on zero frames stripping = on silence cutting = on level thres = 2000 / 32767 zerocross thres = 60 / sec. head margin = 300 msec. tail margin = 400 msec. remove DC offset = off reject short input = off short pause segmentation= off result output to = tty (standard out) output charset conv. = disabled ------------- System Info end ------------- ************************************************************ * NOTICE: The first input may not be correctly recoginized * * since no CMN parameter is available on startup. * ************************************************************ ------ ### read waveform input <<< please speak >>> |