Go to the documentation of this file. 108 #ifndef _CMDLN_MACRO_H_ 109 #define _CMDLN_MACRO_H_ 129 #define vq_cluster_command_line_macro() \ 133 "Use std.dev. (rather than var) in computing vector distances during clustering" }, \ 137 "Stopping criterion: stop iterations if relative decrease in sq(error) < eps" }, \ 141 "Max no. of k-means iterations for clustering" } 143 #define gmm_command_line_macro() \ 147 "Mixture gaussian means input file" },\ 151 "Mixture gaussian variances input file" },\ 155 "Mixture gaussian variance floor (applied to data from -var file)" },\ 159 "Senone mixture weights input file" },\ 163 "Senone mixture weights floor (applied to data from -mixw file)" } 165 #define acoustic_model_command_line_macro() \ 166 gmm_command_line_macro(), \ 170 "Directory for specifying Sphinx 3's hmm, the following files are assummed to be present, mdef, mean, var, mixw, tmat. If -mdef, -mean, -var, -mixw or -tmat are specified, they will override this command. "}, \ 174 "File containing feature extraction parameters."}, \ 178 "Model definition input file" },\ 182 "HMM state transition matrix input file" },\ 186 "HMM state transition probability floor (applied to -tmat file)" },\ 190 "Senone to mixture-gaussian mapping file (or .semi. or .cont.)" }, \ 194 "(S3.0 GMM Computation only) No. of top scoring densities computed in each mixture gaussian codebook (semi-continuous models only)" } 196 #define language_model_command_line_macro() \ 200 "Word trigram language model input file" }, \ 204 "Specify a set of language model\n"}, \ 208 "The directory for dumping the DMP file. "}, \ 212 "Name of language model in -lmctlfn to use for all utterances" } 219 "A finite state grammar control file" },
222 #define finite_state_grammar_command_line_macro() \ 226 "(FSG Mode (Mode 2) only) Finite state grammar"}, \ 227 { "-fsgusealtpron", \ 230 "(FSG Mode (Mode 2) only) Use alternative pronunciations for FSG"}, \ 234 "(FSG Mode (Mode 2) only) Insert filler words at each state."} 237 #define log_table_command_line_macro() \ 241 "Base in which all log-likelihoods calculated" }, \ 245 "Determines whether to use the logs3 table or to compute the values at run time."} 247 #define phoneme_lookahead_command_line_macro() \ 251 "0 = bypass, 1= sum of max, 2 = sum of avg, 3 = sum of 1st senones only" }, \ 255 "Window size (actually window size-1) of phoneme look-ahead." }, \ 259 "Beam for phoneme look-ahead. [1 (narrowest)..10000000(very wide)]" } 261 #define histogram_pruning_command_line_macro() \ 265 "(Only used in Mode 4 and 5) Max no. of distinct word exits to maintain at each frame" }, \ 269 "(Only used in Mode 4 and 5) Max no. of histories to maintain at each frame" }, \ 270 { "-hmmhistbinsize", \ 273 "(Only used in Mode 4 and 5) Performance histogram: #frames vs #HMMs active; #HMMs/bin in this histogram" }, \ 277 "(Only used in Mode 4 and 5) Max no. of active HMMs to maintain at each frame; approx." } 279 #define dictionary_command_line_macro() \ 283 "Main pronunciation dictionary (lexicon) input file" }, \ 287 "Silence and filler (noise) word pronunciation dictionary input file" }, \ 291 "Use CMUDict letter-to-sound rules to generate pronunciations for LM words doesn't appear in the dictionary . Use it with care. It assumes that the phone set in the mdef and dict are the same as the LTS rule. "} 293 #define gaussian_selection_command_line_macro() \ 297 "Gaussian Selection Mapping." } 299 #define fast_GMM_computation_command_line_macro() \ 303 "Sub-vector quantized form of acoustic model" }, \ 307 "Beam selecting best components within each mixture Gaussian [0(widest)..1(narrowest)]" }, \ 308 gaussian_selection_command_line_macro(), \ 312 "Ratio of Down-sampling the frame computation." }, \ 316 "Conditional Down-sampling, override normal down sampling. require specify a gaussian selection map" }, \ 320 "Distance-based Down-sampling, override normal down sampling." }, \ 324 "A flag that specified whether the input GS map will be used for Gaussian Selection. If it is disabled, the map will only provide information to other modules." }, \ 328 "A flag that specified whether the input SVQ will be used as approximate scores of the Gaussians" }, \ 332 "CI phone beam for CI-based GMM Selection. [0(widest) .. 1(narrowest)]"}, \ 337 "From 0 to 1, it tightens the beam width when the frame is dropped"}, \ 341 "Max no. of distinct CD senone will be computed. " }, \ 345 "Number of subvectors to use for SubVQ-based frame evaluation (3 for all)"}, \ 349 "kd-Tree file for Gaussian selection (for .s2semi models only)" }, \ 353 "Maximum depth of kd-Trees to use" }, \ 357 "Maximum number of Gaussians per leaf node in kd-Trees" } 363 "Feature stream:\n\t\t\t\ts2_4x: Sphinx-II type 4 streams, 12cep, 24dcep, 3pow, 12ddcep\n\t\t\t\ts3_1x39: Single stream, 12cep+12dcep+3pow+12ddcep\n\t\t\t\t1s_12c_12d_3p_12dd: Single stream, 12cep+12dcep+3pow+12ddcep\n\t\t\t\t1s_c: Single stream, given input vector only\n\t\t\t\t1s_c_d: Feature + Deltas only\n\t\t\t\t1s_c_dd: Feature + Double deltas only\n\t\t\t\t1s_c_d_dd: Feature + Deltas + Double deltas\n\t\t\t\t1s_c_wd_dd: Feature cep+windowed delcep+deldel \n\t\t\t1s_c_d_ld_dd: Feature + delta + longter delta + doubledelta" },
368 "Feature stream: s2_4x / s3_1x39 / cep_dcep[,%d] / cep[,%d] / %d,%d,...,%d" },
372 #define speaker_adaptation_command_line_macro() \ 376 "MLLR transfomation matrix to be applied to mixture gaussian means"}, \ 380 "Senone to MLLR transformation matrix mapping file (or .1cls.)" } 383 #define common_filler_properties_command_line_macro() \ 387 "Filler word probabilities input file (used in place of -silpen and -noisepen)" }, \ 391 "Default silence word probability" }, \ 395 "Default non-silence filler word probability" }, \ 399 "Language weight" }, \ 403 "Word insertion penalty" }, \ 411 #define phone_insertion_penalty_command_line_macro() \ 415 "(Mode 2 and 3 only) Word insertion penalty" } 418 #define common_s3x_beam_properties_command_line_macro() \ 422 "Beam selecting active HMMs (relative to best) in each frame [0(widest)..1(narrowest)]" }, \ 426 "Beam selecting HMMs transitioning to successors in each frame [0(widest)..1(narrowest)]" }, \ 430 "Beam selecting word-final HMMs exiting in each frame [0(widest)..1(narrowest)]" }, \ 434 "Beam selecting word-final HMMs exiting in each frame [0(widest) .. 1(narrowest)]" }, \ 438 "(Not used in Mode 3) Use wbeam for phone transitions every so many frames (if >= 1)" } 440 #define common_application_properties_command_line_macro() \ 444 "Log file (default stdout/stderr)" } 446 #define control_file_handling_command_line_macro() \ 450 "Control file listing utterances to be processed" }, \ 454 "No. of utterances at the beginning of -ctl file to be skipped" }, \ 458 "No. of utterances to be processed (after skipping -ctloffset entries)" } 460 #define hypothesis_file_handling_command_line_macro() \ 464 "Recognition result file, with only words" }, \ 468 "Recognition result file, with word segmentations and scores" } 470 #define score_handling_command_line_macro() \ 471 { "-hypsegscore_unscale", \ 474 "When displaying the results, whether to unscale back the acoustic score with the best score in a frame"} 476 #define cepstral_input_handling_command_line_macro() \ 480 "Input cepstrum files directory (prefixed to filespecs in control file)" }, \ 484 "Input cepstrum files extension (prefixed to filespecs in control file)" }, \ 488 "Input is waveform data rather than cepstra (-cepdir and -cepext are still used)" }, \ 492 "Number of bytes to skip at the beginning of a waveform file (44 for WAV, 1024 for Sphere)" } 494 #define output_lattice_handling_command_line_macro() \ 498 "Directory in which to dump word lattices" }, \ 502 "Format in which to dump word lattices (either 's3' or 'htk')" }, \ 506 "Filename extension for lattice files (gzip compressed, by default - remove .gz for uncompressed)" } 510 #define history_table_command_line_macro() \ 514 "Directory in which to dump word Viterbi back pointer table (for debugging)" }, \ 518 "Number of BPtable entries to allocate initially (grown as necessary)" } 523 #define decode_specific_command_line_macro() \ 527 "Decoding mode, one of allphone, fsg, fwdflat, fwdtree."}, \ 531 "Operation mode, for internal use only."}, \ 535 "Whether to dump active HMM details to stderr (for debugging)" }, \ 539 "Whether to dump the lextree structure to stderr (for debugging), 1 for Ravi's format, 2 for Dot format, Larger than 2 will be treated as Ravi's format" }, \ 543 "Bigram-mode: If TRUE only one BP entry/frame; else one per LM state" }, \ 547 "If true, Use unigram probs in lextree" } 549 #define dag_handling_command_line_macro() \ 553 "Nodes ignored during search if they persist for fewer than so many end frames" }, \ 557 "(0..2); 1 or 2: add edge if endframe == startframe; 2: if start == end-1" }, \ 561 "Max DAG edges allowed in utterance; aborted if exceeded; controls memory usage" }, \ 565 "Max LMops in utterance after which it is aborted; controls CPU use (see maxlpf)" }, \ 569 "Max LMops/frame after which utterance aborted; controls CPU use (see maxlmop)" }, \ 573 "Whether lattice is compressed."} 576 #define second_stage_dag_handling_command_line_macro() \ 580 "Whether to run bestpath DAG search after forward Viterbi pass" }, \ 584 "Language weight for bestpath DAG search (default: same as -lw)" }, \ 588 "Input word-lattice directory with per-utt files for restricting words searched"}, \ 592 "N-best filename extension (.gz or .Z extension for compression)"}, \ 596 "Max. n-best hypotheses to generate per utterance"}, \ 600 "Max partial paths created after which utterance aborted; controls CPU/memory use"}, \ 604 "Generate debugging information for N-best search. "} 606 #define input_lattice_handling_command_line_macro() \ 610 "Input word-lattice directory with per-utt files for restricting words searched" }, \ 614 "Input word-lattice words starting within +/- <this argument> of current frame considered during search" } 616 #define flat_fwd_debugging_command_line_macro() \ 620 "(Mode 3 only) Word whose active HMMs are to be traced (for debugging/diagnosis/analysis)" }, \ 624 "(Mode 3 only) Ending frame for dumping all active HMMs (for debugging/diagnosis/analysis)" }, \ 628 "(Mode 3 only) Starting frame for dumping all active HMMs (for debugging/diagnosis/analysis)" }, \ 632 "(Mode 3 only) Ending frame for dumping all active words (for debugging/diagnosis/analysis)" }, \ 636 "(Mode 3 only) Starting frame for dumping all active words (for debugging/diagnosis/analysis)" } 639 #define search_specific_command_line_macro() \ 643 "Whether detailed backtrace information (word segmentation/scores) shown in log" }, \ 644 { "-bestsenscrdir", \ 647 "When Best senone score directory." } 651 #define search_modeTST_specific_command_line_macro() \ 655 "(Mode 4 only) No. of lextrees to be instantiated; entries into them staggered in time" }, \ 659 "(Mode 4 only) Entries Per Lextree; #successive entries into one lextree before lextree-entries shifted to the next" } 662 #define search_modeWST_specific_command_line_macro() \ 666 "(Mode 5 only) No. of lextrees to be instantiated statically; " } 668 #define partial_hypothesis_command_line_macro() \ 672 "(Live-decoder only) Maximum number of words in a partial hypothesis (for block decoding)" }, \ 676 "(Live-decoder only) dump parital hypothesis on the screen"} 678 #define control_lm_file_command_line_macro() \ 682 "(Not used in mode 2 and 3) Control file that list the corresponding LMs" } 684 #define control_mllr_file_command_line_macro() \ 688 "Control file that list the corresponding MLLR matrix for an utterance"} 690 #define control_lm_mllr_file_command_line_macro() \ 691 control_lm_file_command_line_macro(), \ 692 control_mllr_file_command_line_macro()