INFO: cmn.c(133): CMN: 55.15 11.33 0.66 17.85 -5.40 12.36 -5.03 11.33 2.66 2.03 -11.61 -6.28 4.88
INFO: ngram_search.c(459): Resized backpointer table to 10000 entries
INFO: ngram_search_fwdtree.c(1550): 9786 words recognized (12/fr)
INFO: ngram_search_fwdtree.c(1552): 134479 senones evaluated (167/fr)
INFO: ngram_search_fwdtree.c(1556): 1267419 channels searched (1578/fr), 417642 1st, 661157 last
INFO: ngram_search_fwdtree.c(1559): 276004 words for which last channels evaluated (343/fr)
INFO: ngram_search_fwdtree.c(1561): 359195 candidate words for entering last phone (447/fr)
INFO: ngram_search_fwdtree.c(1564): fwdtree 0.81 CPU 0.101 xRT
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.81 wall 0.101 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 434 words
INFO: ngram_search_fwdflat.c(948): 7299 words recognized (9/fr)
INFO: ngram_search_fwdflat.c(950): 46878 senones evaluated (58/fr)
INFO: ngram_search_fwdflat.c(952): 99859 channels searched (124/fr)
INFO: ngram_search_fwdflat.c(954): 37490 words searched (46/fr)
INFO: ngram_search_fwdflat.c(957): 31099 word transitions (38/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.14 CPU 0.017 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.14 wall 0.017 xRT
INFO: ngram_search.c(1250): lattice start node .0 end node .781
INFO: ngram_search.c(1276): Eliminated 0 nodes before end node
INFO: ngram_search.c(1381): Lattice has 696 nodes, 4565 links
INFO: ps_lattice.c(1380): Bestpath score: -28081
INFO: ps_lattice.c(1384): Normalizer P(O) = alpha(:781:801) = -2095355
INFO: ps_lattice.c(1441): Joint P(O,S) = -2216742 P(S|O) = -121387
INFO: ngram_search.c(872): bestpath 0.02 CPU 0.002 xRT
INFO: ngram_search.c(875): bestpath 0.02 wall 0.003 xRT
Can someone explain to me what each of these outputs mean as there is no documentation for pocketsphinx debug output... Though i know how the decoding works I'am not able to interpret the output
Thanks in advance
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
INFO: cmn.c(133): CMN: 55.15 11.33 0.66 17.85 -5.40 12.36 -5.03 11.33 2.66 2.03 -11.61 -6.28 4.88
INFO: ngram_search.c(459): Resized backpointer table to 10000 entries
INFO: ngram_search_fwdtree.c(1550): 9786 words recognized (12/fr)
INFO: ngram_search_fwdtree.c(1552): 134479 senones evaluated (167/fr)
INFO: ngram_search_fwdtree.c(1556): 1267419 channels searched (1578/fr), 417642 1st, 661157 last
INFO: ngram_search_fwdtree.c(1559): 276004 words for which last channels evaluated (343/fr)
INFO: ngram_search_fwdtree.c(1561): 359195 candidate words for entering last phone (447/fr)
INFO: ngram_search_fwdtree.c(1564): fwdtree 0.81 CPU 0.101 xRT
INFO: ngram_search_fwdtree.c(1567): fwdtree 0.81 wall 0.101 xRT
INFO: ngram_search_fwdflat.c(302): Utterance vocabulary contains 434 words
INFO: ngram_search_fwdflat.c(948): 7299 words recognized (9/fr)
INFO: ngram_search_fwdflat.c(950): 46878 senones evaluated (58/fr)
INFO: ngram_search_fwdflat.c(952): 99859 channels searched (124/fr)
INFO: ngram_search_fwdflat.c(954): 37490 words searched (46/fr)
INFO: ngram_search_fwdflat.c(957): 31099 word transitions (38/fr)
INFO: ngram_search_fwdflat.c(960): fwdflat 0.14 CPU 0.017 xRT
INFO: ngram_search_fwdflat.c(963): fwdflat 0.14 wall 0.017 xRT
INFO: ngram_search.c(1250): lattice start node
.0 end node.781INFO: ngram_search.c(1276): Eliminated 0 nodes before end node
INFO: ngram_search.c(1381): Lattice has 696 nodes, 4565 links
INFO: ps_lattice.c(1380): Bestpath score: -28081
INFO: ps_lattice.c(1384): Normalizer P(O) = alpha(:781:801) = -2095355
INFO: ps_lattice.c(1441): Joint P(O,S) = -2216742 P(S|O) = -121387
INFO: ngram_search.c(872): bestpath 0.02 CPU 0.002 xRT
INFO: ngram_search.c(875): bestpath 0.02 wall 0.003 xRT
Can someone explain to me what each of these outputs mean as there is no documentation for pocketsphinx debug output... Though i know how the decoding works I'am not able to interpret the output
Thanks in advance
There is a documentation:
http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.72.3560
You can find answers for your questions there.