问题：95％的时间我没有得到任何结果，其余的只是来自pocketsphinx的准确单词。
这可能是录音量低吗？
到目前为止：
Pocketsphinx初始化时包含默认的hmm，lm和dict。设置代码是:(没有崩溃，没有任何问题）。
g_NPCController.Debug("Initializing internal decoder ... ");
string directoryPrefix = Directory.GetCurrentDirectory() + Path.DirectorySeparatorChar +
"Pocketsphinx" + Path.DirectorySeparatorChar + "model";
string hmmDir = directoryPrefix + Path.DirectorySeparatorChar + "en-us" + Path.DirectorySeparatorChar + "en-us";
string dictDir = directoryPrefix + Path.DirectorySeparatorChar + "en-us" + Path.DirectorySeparatorChar + "cmudict-en-us.dict";
string lmDir = directoryPrefix + Path.DirectorySeparatorChar + "en-us" + Path.DirectorySeparatorChar + "en-us.lm.bin";
Config c = Pocketsphinx.Decoder.DefaultConfig();
if (Application.platform == RuntimePlatform.Android)
{
c.SetString("-hmm", "/sdcard/Android/data/edu.cmu.sphinx.pocketsphinx/files/sync/en-us-ptm");
c.SetString("-dict", "/sdcard/Android/data/edu.cmu.sphinx.pocketsphinx/files/sync/cmudict-en-us.dict");
c.SetString("-lm", "/sdcard/Android/data/edu.cmu.sphinx.pocketsphinx/files/sync/en-us.lm.bin");
} else if (Application.platform == RuntimePlatform.IPhonePlayer) {
// TODO - outta luck for now
}
else
{
if (g_NPCController.DebugMode)
c.SetString("-logfn", Directory.GetCurrentDirectory() + Path.DirectorySeparatorChar + "Pocketsphinx" + Path.DirectorySeparatorChar + "current.log");
c.SetString("-hmm", hmmDir);
c.SetString("-dict", dictDir);
c.SetString("-lm", lmDir);
}
//c.SetString("-keyphrase", "hello world");
c.SetFloat("-kws_threshold", 1e-30);
c.SetFloat("-samprate", (int) g_NPCAudioListener.SampleFrequency);
c.SetInt("-nfft", 2048);
g_Decoder = new Pocketsphinx.Decoder(c);
g_Decoder.StartUtt();
g_NPCController.Debug("... local decoder initialized.");
然后，在主线程的coroutne中，缓冲区一次处理。缓冲区每个可以是2600到12800字节。每帧调用以下代码：
while (!buffer.Closed)
yield return null;
g_NPCAudioListener.AudioBufferQueue.Dequeue();
byte[] audio = new byte[buffer.CurrentBuffer16.Count * sizeof(short)];
Buffer.BlockCopy(buffer.CurrentBuffer16.ToArray(), 0, audio, 0, audio.Length);
g_Decoder.ProcessRaw(audio, audio.Length, false, consumed == buffer.CurrentBuffer.Count);
if (g_Decoder.Hyp() != null) {
g_DictationResults.Enqueue(g_Decoder.Hyp().Hypstr);
g_Decoder.EndUtt();
g_Decoder.StartUtt();
}
任何帮助将不胜感激。我非常接近使它成功 - 这对我的项目来说是一个巨大的帮助。我错过了一些配置参数吗？
使用init log：
INFO: pocketsphinx.c(152): Parsed model-specific feature parameters from C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\model\en-us\en-us/feat.params
Current configuration:
[NAME] [DEFLT] [VALUE]
-agc none none
-agcthresh 2.0 2.000000e+00
-allphone
-allphone_ci yes yes
-alpha 0.97 9.700000e-01
-ascale 20.0 2.000000e+01
-aw 1 1
-backtrace no no
-beam 1e-48 1.000000e-48
-bestpath yes yes
-bestpathlw 9.5 9.500000e+00
-ceplen 13 13
-cmn live current
-cmninit 40,3,-1 40,3,-1
-compallsen no no
-dict C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\model\en-us\cmudict-en-us.dict
-dictcase no no
-dither no no
-doublebw no no
-ds 1 1
-fdict
-feat 1s_c_d_dd 1s_c_d_dd
-featparams
-fillprob 1e-8 1.000000e-08
-frate 100 100
-fsg
-fsgusealtpron yes yes
-fsgusefiller yes yes
-fwdflat yes yes
-fwdflatbeam 1e-64 1.000000e-64
-fwdflatefwid 4 4
-fwdflatlw 8.5 8.500000e+00
-fwdflatsfwin 25 25
-fwdflatwbeam 7e-29 7.000000e-29
-fwdtree yes yes
-hmm C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\model\en-us\en-us
-input_endian little little
-jsgf
-keyphrase
-kws
-kws_delay 10 10
-kws_plp 1e-1 1.000000e-01
-kws_threshold 1e-30 1.000000e-30
-latsize 5000 5000
-lda
-ldadim 0 0
-lifter 0 22
-lm C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\model\en-us\en-us.lm.bin
-lmctl
-lmname
-logbase 1.0001 1.000100e+00
-logfn C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\current.log
-logspec no no
-lowerf 133.33334 1.300000e+02
-lpbeam 1e-40 1.000000e-40
-lponlybeam 7e-29 7.000000e-29
-lw 6.5 6.500000e+00
-maxhmmpf 30000 30000
-maxwpf -1 -1
-mdef
-mean
-mfclogdir
-min_endfr 0 0
-mixw
-mixwfloor 0.0000001 1.000000e-07
-mllr
-mmap yes yes
-ncep 13 13
-nfft 512 2048
-nfilt 40 25
-nwpen 1.0 1.000000e+00
-pbeam 1e-48 1.000000e-48
-pip 1.0 1.000000e+00
-pl_beam 1e-10 1.000000e-10
-pl_pbeam 1e-10 1.000000e-10
-pl_pip 1.0 1.000000e+00
-pl_weight 3.0 3.000000e+00
-pl_window 5 5
-rawlogdir
-remove_dc no no
-remove_noise yes yes
-remove_silence yes yes
-round_filters yes yes
-samprate 16000 1.600000e+04
-seed -1 -1
-sendump
-senlogdir
-senmgau
-silprob 0.005 5.000000e-03
-smoothspec no no
-svspec 0-12/13-25/26-38
-tmat
-tmatfloor 0.0001 1.000000e-04
-topn 4 4
-topn_beam 0 0
-toprule
-transform legacy dct
-unit_area yes yes
-upperf 6855.4976 6.800000e+03
-uw 1.0 1.000000e+00
-vad_postspeech 50 50
-vad_prespeech 20 20
-vad_startspeech 10 10
-vad_threshold 3.0 3.000000e+00
-var
-varfloor 0.0001 1.000000e-04
-varnorm no no
-verbose no no
-warp_params
-warp_type inverse_linear inverse_linear
-wbeam 7e-29 7.000000e-29
-wip 0.65 6.500000e-01
-wlen 0.025625 2.562500e-02
INFO: feat.c(715): Initializing feature stream to type: '1s_c_d_dd', ceplen=13, CMN='batch', VARNORM='no', AGC='none'
INFO: acmod.c(162): Using subvector specification 0-12/13-25/26-38
INFO: mdef.c(518): Reading model definition: C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\model\en-us\en-us/mdef
INFO: mdef.c(531): Found byte-order mark BMDF, assuming this is a binary mdef file
INFO: bin_mdef.c(336): Reading binary model definition: C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\model\en-us\en-us/mdef
INFO: bin_mdef.c(516): 42 CI-phone, 137053 CD-phone, 3 emitstate/phone, 126 CI-sen, 5126 Sen, 29324 Sen-Seq
INFO: tmat.c(149): Reading HMM transition probability matrices: C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\model\en-us\en-us/transition_matrices
INFO: acmod.c(113): Attempting to use PTM computation module
INFO: ms_gauden.c(127): Reading mixture gaussian parameter: C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\model\en-us\en-us/means
INFO: ms_gauden.c(242): 42 codebook, 3 feature, size:
INFO: ms_gauden.c(244): 128x13
INFO: ms_gauden.c(244): 128x13
INFO: ms_gauden.c(244): 128x13
INFO: ms_gauden.c(127): Reading mixture gaussian parameter: C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\model\en-us\en-us/variances
INFO: ms_gauden.c(242): 42 codebook, 3 feature, size:
INFO: ms_gauden.c(244): 128x13
INFO: ms_gauden.c(244): 128x13
INFO: ms_gauden.c(244): 128x13
INFO: ms_gauden.c(304): 222 variance values floored
INFO: ptm_mgau.c(475): Loading senones from dump file C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\model\en-us\en-us/sendump
INFO: ptm_mgau.c(499): BEGIN FILE FORMAT DESCRIPTION
INFO: ptm_mgau.c(562): Rows: 128, Columns: 5126
INFO: ptm_mgau.c(594): Using memory-mapped I/O for senones
INFO: ptm_mgau.c(837): Maximum top-N: 4
INFO: phone_loop_search.c(114): State beam -225 Phone exit beam -225 Insertion penalty 0
INFO: dict.c(320): Allocating 138623 * 32 bytes (4331 KiB) for word entries
INFO: dict.c(333): Reading main dictionary: C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\model\en-us\cmudict-en-us.dict
INFO: dict.c(213): Dictionary size 134522, allocated 1014 KiB for strings, 1677 KiB for phones
INFO: dict.c(336): 134522 words read
INFO: dict.c(358): Reading filler dictionary: C:\Users\fgera\Development\Git\Computer-Graphics\Motional.AI\Pocketsphinx\model\en-us\en-us/noisedict
INFO: dict.c(213): Dictionary size 134527, allocated 0 KiB for strings, 0 KiB for phones
INFO: dict.c(361): 5 words read
INFO: dict2pid.c(396): Building PID tables for dictionary
INFO: dict2pid.c(406): Allocating 42^3 * 2 bytes (144 KiB) for word-initial triphones
INFO: dict2pid.c(132): Allocated 42672 bytes (41 KiB) for word-final triphones
INFO: dict2pid.c(196): Allocated 42672 bytes (41 KiB) for single-phone word triphones
INFO: ngram_model_trie.c(354): Trying to read LM in trie binary format
INFO: ngram_search_fwdtree.c(74): Initializing search tree
INFO: ngram_search_fwdtree.c(101): 790 unique initial diphones
INFO: ngram_search_fwdtree.c(186): Creating search channels
INFO: ngram_search_fwdtree.c(323): Max nonroot chan increased to 152144
INFO: ngram_search_fwdtree.c(333): Created 722 root, 152016 non-root channels, 53 single-phone words
INFO: ngram_search_fwdflat.c(157): fwdflat: min_ef_width = 4, max_sf_win = 25
INFO: cmn_live.c(120): Update from < 40.00 3.00 -1.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 0.00 >
INFO: cmn_live.c(138): Update to < 95.43 -6.79 -5.30 -6.10 -10.80 2.20 -1.31 -0.54 -2.47 -3.11 -0.83 2.18 5.92 >
INFO: ngram_search_fwdtree.c(1550): 1601 words recognized (18/fr)
INFO: ngram_search_fwdtree.c(1552): 237595 senones evaluated (2700/fr)
INFO: ngram_search_fwdtree.c(1556): 578857 channels searched (6577/fr), 49670 1st, 54325 last
INFO: ngram_search_fwdtree.c(1559): 3042 words for which last channels evaluated (34/fr)
INFO: ngram_search_fwdtree.c(1561): 18162 candidate words for entering last phone (206/fr)
INFO: ngram_search_fwdtree.c(1564): fwdtree 18.25 CPU 20.739 xRT
INFO: ngram_search_fwdtree.c(1567): fwdtree 14.14 wall 16.065 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 57 words
INFO: ngram_search_fwdflat.c(948): 692 words recognized (8/fr)
INFO: ngram_search_fwdflat.c(950): 83914 senones evaluated (954/fr)
INFO: ngram_search_fwdflat.c(952): 102603 channels searched (1165/fr)
INFO: ngram_search_fwdflat.c(954): 4800 words searched (54/fr)
INFO: ngram_search_fwdflat.c(957): 2171 word transitions (24/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.03 CPU 0.036 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.03 wall 0.033 xRT
INFO: cmn_live.c(120): Update from < 95.43 -6.79 -5.30 -6.10 -10.80 2.20 -1.31 -0.54 -2.47 -3.11 -0.83 2.18 5.92 >
INFO: cmn_live.c(138): Update to < 92.77 -8.91 -9.40 -6.80 -11.46 -0.71 -2.89 -0.45 1.43 -3.17 -1.35 0.17 3.64 >
INFO: ngram_search_fwdtree.c(1550): 4191 words recognized (37/fr)
INFO: ngram_search_fwdtree.c(1552): 451300 senones evaluated (4029/fr)
INFO: ngram_search_fwdtree.c(1556): 3011851 channels searched (26891/fr), 76495 1st, 179109 last
INFO: ngram_search_fwdtree.c(1559): 8886 words for which last channels evaluated (79/fr)
INFO: ngram_search_fwdtree.c(1561): 234585 candidate words for entering last phone (2094/fr)
INFO: ngram_search_fwdtree.c(1564): fwdtree 28.91 CPU 25.809 xRT
INFO: ngram_search_fwdtree.c(1567): fwdtree 18.27 wall 16.315 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 167 words
INFO: ngram_search_fwdflat.c(948): 2425 words recognized (22/fr)
INFO: ngram_search_fwdflat.c(950): 166879 senones evaluated (1490/fr)
INFO: ngram_search_fwdflat.c(952): 279618 channels searched (2496/fr)
INFO: ngram_search_fwdflat.c(954): 13084 words searched (116/fr)
INFO: ngram_search_fwdflat.c(957): 9930 word transitions (88/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.08 CPU 0.070 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.08 wall 0.074 xRT
INFO: ngram_search.c(467): Resized score stack to 200000 entries
INFO: ngram_search.c(459): Resized backpointer table to 10000 entries
INFO: cmn_live.c(120): Update from < 92.77 -8.91 -9.40 -6.80 -11.46 -0.71 -2.89 -0.45 1.43 -3.17 -1.35 0.17 3.64 >
INFO: cmn_live.c(138): Update to < 93.21 -8.67 -8.88 -5.56 -10.87 -0.19 -2.78 -0.73 2.84 -3.07 -1.93 0.79 2.44 >
INFO: ngram_search_fwdtree.c(1550): 6212 words recognized (97/fr)
INFO: ngram_search_fwdtree.c(1552): 251244 senones evaluated (3926/fr)
INFO: ngram_search_fwdtree.c(1556): 2124350 channels searched (33192/fr), 43302 1st, 190124 last
INFO: ngram_search_fwdtree.c(1559): 9693 words for which last channels evaluated (151/fr)
INFO: ngram_search_fwdtree.c(1561): 212658 candidate words for entering last phone (3322/fr)
INFO: ngram_search_fwdtree.c(1564): fwdtree 13.48 CPU 21.069 xRT
INFO: ngram_search_fwdtree.c(1567): fwdtree 9.23 wall 14.425 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 311 words
INFO: ngram_search_fwdflat.c(948): 3854 words recognized (60/fr)
INFO: ngram_search_fwdflat.c(950): 138888 senones evaluated (2170/fr)
INFO: ngram_search_fwdflat.c(952): 414216 channels searched (6472/fr)
INFO: ngram_search_fwdflat.c(954): 18404 words searched (287/fr)
INFO: ngram_search_fwdflat.c(957): 11245 word transitions (175/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.14 CPU 0.220 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.12 wall 0.187 xRT
INFO: cmn_live.c(120): Update from < 93.21 -8.67 -8.88 -5.56 -10.87 -0.19 -2.78 -0.73 2.84 -3.07 -1.93 0.79 2.44 >
INFO: cmn_live.c(138): Update to < 93.41 -9.57 -8.47 -5.47 -10.06 0.42 -2.95 -1.82 2.57 -2.62 -0.98 1.79 2.24 >
INFO: ngram_search_fwdtree.c(1550): 3858 words recognized (54/fr)
INFO: ngram_search_fwdtree.c(1552): 236795 senones evaluated (3289/fr)
INFO: ngram_search_fwdtree.c(1556): 1169796 channels searched (16247/fr), 40281 1st, 142488 last
INFO: ngram_search_fwdtree.c(1559): 7030 words for which last channels evaluated (97/fr)
INFO: ngram_search_fwdtree.c(1561): 72648 candidate words for entering last phone (1009/fr)
INFO: ngram_search_fwdtree.c(1564): fwdtree 17.91 CPU 24.870 xRT
INFO: ngram_search_fwdtree.c(1567): fwdtree 12.32 wall 17.115 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 155 words
INFO: ngram_search_fwdflat.c(948): 2938 words recognized (41/fr)
INFO: ngram_search_fwdflat.c(950): 111642 senones evaluated (1551/fr)
INFO: ngram_search_fwdflat.c(952): 229547 channels searched (3188/fr)
INFO: ngram_search_fwdflat.c(954): 10456 words searched (145/fr)
INFO: ngram_search_fwdflat.c(957): 6314 word transitions (87/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.08 CPU 0.109 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.08 wall 0.104 xRT
INFO: cmn_live.c(120): Update from < 93.41 -9.57 -8.47 -5.47 -10.06 0.42 -2.95 -1.82 2.57 -2.62 -0.98 1.79 2.24 >
INFO: cmn_live.c(138): Update to < 93.74 -10.48 -8.87 -5.63 -9.42 0.32 -3.21 -2.07 2.21 -1.54 -0.68 1.97 1.85 >
INFO: ngram_search_fwdtree.c(1550): 7019 words recognized (80/fr)
INFO: ngram_search_fwdtree.c(1552): 325095 senones evaluated (3694/fr)
INFO: ngram_search_fwdtree.c(1556): 1955385 channels searched (22220/fr), 58503 1st, 232252 last
INFO: ngram_search_fwdtree.c(1559): 11334 words for which last channels evaluated (128/fr)
INFO: ngram_search_fwdtree.c(1561): 115217 candidate words for entering last phone (1309/fr)
INFO: ngram_search_fwdtree.c(1564): fwdtree 18.08 CPU 20.543 xRT
INFO: ngram_search_fwdtree.c(1567): fwdtree 11.98 wall 13.615 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 365 words
INFO: ngram_search_fwdflat.c(948): 2632 words recognized (30/fr)
INFO: ngram_search_fwdflat.c(950): 176368 senones evaluated (2004/fr)
INFO: ngram_search_fwdflat.c(952): 522457 channels searched (5937/fr)
INFO: ngram_search_fwdflat.c(954): 23759 words searched (269/fr)
INFO: ngram_search_fwdflat.c(957): 13859 word transitions (157/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.11 CPU 0.124 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.11 wall 0.130 xRT
INFO: cmn_live.c(120): Update from < 93.74 -10.48 -8.87 -5.63 -9.42 0.32 -3.21 -2.07 2.21 -1.54 -0.68 1.97 1.85 >
INFO: cmn_live.c(138): Update to < 93.53 -10.03 -8.85 -4.80 -8.58 0.32 -3.52 -2.14 2.91 -1.47 -0.63 2.22 1.90 >
INFO: ngram_search_fwdtree.c(1550): 883 words recognized (21/fr)
INFO: ngram_search_fwdtree.c(1552): 146888 senones evaluated (3416/fr)
INFO: ngram_search_fwdtree.c(1556): 750300 channels searched (17448/fr), 25850 1st, 46629 last
INFO: ngram_search_fwdtree.c(1559): 2323 words for which last channels evaluated (54/fr)
INFO: ngram_search_fwdtree.c(1561): 53709 candidate words for entering last phone (1249/fr)
INFO: ngram_search_fwdtree.c(1564): fwdtree 17.42 CPU 40.516 xRT
INFO: ngram_search_fwdtree.c(1567): fwdtree 12.03 wall 27.979 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 53 words
INFO: ngram_search_fwdflat.c(948): 650 words recognized (15/fr)
INFO: ngram_search_fwdflat.c(950): 37862 senones evaluated (881/fr)
INFO: ngram_search_fwdflat.c(952): 45609 channels searched (1060/fr)
INFO: ngram_search_fwdflat.c(954): 2226 words searched (51/fr)
INFO: ngram_search_fwdflat.c(957): 1814 word transitions (42/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.02 CPU 0.036 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.02 wall 0.042 xRT
INFO: cmn_live.c(120): Update from < 93.53 -10.03 -8.85 -4.80 -8.58 0.32 -3.52 -2.14 2.91 -1.47 -0.63 2.22 1.90 >
INFO: cmn_live.c(138): Update to < 92.82 -9.69 -8.69 -5.19 -8.84 0.28 -2.89 -2.53 2.95 -0.77 0.05 2.61 1.68 >
INFO: ngram_search_fwdtree.c(1550): 868 words recognized (19/fr)
INFO: ngram_search_fwdtree.c(1552): 165478 senones evaluated (3597/fr)
INFO: ngram_search_fwdtree.c(1556): 1148658 channels searched (24970/fr), 30324 1st, 29845 last
INFO: ngram_search_fwdtree.c(1559): 1763 words for which last channels evaluated (38/fr)
INFO: ngram_search_fwdtree.c(1561): 103611 candidate words for entering last phone (2252/fr)
INFO: ngram_search_fwdtree.c(1564): fwdtree 12.61 CPU 27.412 xRT
INFO: ngram_search_fwdtree.c(1567): fwdtree 8.01 wall 17.409 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 46 words
INFO: ngram_search_fwdflat.c(948): 585 words recognized (13/fr)
INFO: ngram_search_fwdflat.c(950): 35815 senones evaluated (779/fr)
INFO: ngram_search_fwdflat.c(952): 37653 channels searched (818/fr)
INFO: ngram_search_fwdflat.c(954): 2052 words searched (44/fr)
INFO: ngram_search_fwdflat.c(957): 1642 word transitions (35/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.02 CPU 0.034 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.02 wall 0.043 xRT
INFO: cmn_live.c(120): Update from < 92.82 -9.69 -8.69 -5.19 -8.84 0.28 -2.89 -2.53 2.95 -0.77 0.05 2.61 1.68 >
INFO: cmn_live.c(138): Update to < 93.42 -9.82 -8.43 -4.88 -8.43 0.13 -2.56 -2.49 3.26 -0.28 0.04 2.72 1.43 >
INFO: ngram_search_fwdtree.c(1550): 6952 words recognized (67/fr)
INFO: ngram_search_fwdtree.c(1552): 414969 senones evaluated (3990/fr)
INFO: ngram_search_fwdtree.c(1556): 2748306 channels searched (26426/fr), 71316 1st, 227747 last
INFO: ngram_search_fwdtree.c(1559): 11669 words for which last channels evaluated (112/fr)
INFO: ngram_search_fwdtree.c(1561): 197819 candidate words for entering last phone (1902/fr)
INFO: ngram_search_fwdtree.c(1564): fwdtree 14.70 CPU 14.138 xRT
INFO: ngram_search_fwdtree.c(1567): fwdtree 10.16 wall 9.771 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 267 words
INFO: ngram_search_fwdflat.c(948): 4301 words recognized (41/fr)
INFO: ngram_search_fwdflat.c(950): 215433 senones evaluated (2071/fr)
INFO: ngram_search_fwdflat.c(952): 503454 channels searched (4840/fr)
INFO: ngram_search_fwdflat.c(954): 22257 words searched (214/fr)
INFO: ngram_search_fwdflat.c(957): 14100 word transitions (135/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.16 CPU 0.150 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.15 wall 0.146 xRT
INFO: cmn_live.c(120): Update from < 93.42 -9.82 -8.43 -4.88 -8.43 0.13 -2.56 -2.49 3.26 -0.28 0.04 2.72 1.43 >
INFO: cmn_live.c(138): Update to < 93.27 -9.95 -7.95 -3.79 -7.90 0.07 -2.63 -2.50 3.61 -0.30 -0.63 2.60 1.69 >
INFO: ngram_search_fwdtree.c(1550): 1560 words recognized (32/fr)
INFO: ngram_search_fwdtree.c(1552): 174992 senones evaluated (3646/fr)
INFO: ngram_search_fwdtree.c(1556): 1340415 channels searched (27925/fr), 30752 1st, 70997 last
INFO: ngram_search_fwdtree.c(1559): 3334 words for which last channels evaluated (69/fr)
INFO: ngram_search_fwdtree.c(1561): 135428 candidate words for entering last phone (2821/fr)
INFO: ngram_search_fwdtree.c(1564): fwdtree 5.31 CPU 11.068 xRT
INFO: ngram_search_fwdtree.c(1567): fwdtree 3.83 wall 7.975 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 99 words
INFO: ngram_search_fwdflat.c(948): 1259 words recognized (26/fr)
INFO: ngram_search_fwdflat.c(950): 61872 senones evaluated (1289/fr)
INFO: ngram_search_fwdflat.c(952): 107099 channels searched (2231/fr)
INFO: ngram_search_fwdflat.c(954): 4479 words searched (93/fr)
INFO: ngram_search_fwdflat.c(957): 3862 word transitions (80/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.05 CPU 0.098 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.04 wall 0.083 xRT
INFO: cmn_live.c(120): Update from < 93.27 -9.95 -7.95 -3.79 -7.90 0.07 -2.63 -2.50 3.61 -0.30 -0.63 2.60 1.69 >
INFO: cmn_live.c(138): Update to < 93.46 -10.27 -7.86 -3.87 -7.63 -0.25 -2.72 -2.36 3.36 -0.15 -0.46 2.59 1.64 >
INFO: ngram_search_fwdtree.c(1550): 3794 words recognized (53/fr)
INFO: ngram_search_fwdtree.c(1552): 239815 senones evaluated (3331/fr)
INFO: ngram_search_fwdtree.c(1556): 1394923 channels searched (19373/fr), 44549 1st, 127514 last
INFO: ngram_search_fwdtree.c(1559): 6614 words for which last channels evaluated (91/fr)
INFO: ngram_search_fwdtree.c(1561): 107867 candidate words for entering last phone (1498/fr)
INFO: ngram_search_fwdtree.c(1564): fwdtree 23.55 CPU 32.704 xRT
INFO: ngram_search_fwdtree.c(1567): fwdtree 15.71 wall 21.824 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 174 words
INFO: ngram_search_fwdflat.c(948): 2625 words recognized (36/fr)
INFO: ngram_search_fwdflat.c(950): 112676 senones evaluated (1565/fr)
INFO: ngram_search_fwdflat.c(952): 238159 channels searched (3307/fr)
INFO: ngram_search_fwdflat.c(954): 11272 words searched (156/fr)
INFO: ngram_search_fwdflat.c(957): 6236 word transitions (86/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.08 CPU 0.109 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.08 wall 0.110 xRT
INFO: cmn_live.c(120): Update from < 93.46 -10.27 -7.86 -3.87 -7.63 -0.25 -2.72 -2.36 3.36 -0.15 -0.46 2.59 1.64 >
INFO: cmn_live.c(138): Update to < 93.46 -10.27 -7.86 -3.87 -7.63 -0.25 -2.72 -2.36 3.36 -0.15 -0.46 2.59 1.64 >
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 0 words
INFO: ngram_search_fwdtree.c(429): TOTAL fwdtree 179.72 CPU 24.721 xRT
INFO: ngram_search_fwdtree.c(432): TOTAL fwdtree 121.51 wall 16.713 xRT
INFO: ngram_search_fwdflat.c(176): TOTAL fwdflat 0.75 CPU 0.103 xRT
INFO: ngram_search_fwdflat.c(179): TOTAL fwdflat 0.73 wall 0.100 xRT
INFO: ngram_search.c(303): TOTAL bestpath 0.00 CPU 0.000 xRT
INFO: ngram_search.c(306): TOTAL bestpath 0.00 wall 0.000 xRT