|
SPRAAK
|
All Data Structures Namespaces Files Functions Variables Typedefs Enumerations Enumerator Groups Pages
|
SPRAAK
|
Viterbi alignment. More...
Viterbi alignment.
spr_vitalign [-seg segfname](stdout) [-AS](flag: scores) [-S](flag: state alignment) [-W](flag: word alignment) [-WM](flag: word markers) [-U](flag: unit spelling) [-WS](flag: spelling as name) <-beam beam search info> [-s spelling] <-c Corpus> [-range b:e](0:-1) [-ssp script](SPR_BSS_DEV_NULL) [-obs ObsDir] [-suffix FileSuffix](sam) [-arcd FileName] <-h FileName> [-g FileName] [-sel FileName] [-am_opt Options] [-top_n Number(s)](0) [-rmg rmg_params](no) [-LMout Value](-100) [-NOGS](flag: no gauss sel) [-ci CI-file] [-cd CD-file] [-d Dictionary] [-unwind unwind format] (-i==-obs) (-u==-ci)
| -seg<em>segfname</em><a | name="spr_vitalign.seg" class="el"> File to write the segmentations to. |
| -ASflag | scores Add the segment level acoustic scores as extra information to the segmentation file. |
| -Sflag | state alignment Give a sub-unit based alignment instead of the default unit based segmentation. |
| -Wflag | word alignment Give a word based alignment instead of the default unit based segmentation. |
| -WMflag | word markers Insert word markers in the segmentation file. |
| -Uflag | unit spelling The given spelling (option -s) is on the unit level and not on the word level. |
| -WSflag | spelling as name Use the word spelling as segmentation name. By default the filename is used as spelling. |
| -beam<em>beam | search info Parameters for the beam search algorithm. The extra parameters that can be specified are explained in the manual page of load_search_space() |
| -s<em>spelling</em><a | name="spr_vitalign.s" class="el"> Spelling of the word model(s) or units to align with. |
| -c<em>Corpus</em><a | name="spr_vitalign.c" class="el"> File with corpus entries or segmentations. |
| -range<em>b:e</em><a | name="spr_vitalign.range" class="el"> Optional begin and end entry the corpus/segmentation file. Counting starts at 0. |
| -ssp<em>script</em><a | name="spr_vitalign.ssp" class="el"> The signal processing script used to preprocess the input data. |
| -obs<em>ObsDir</em><a | name="spr_vitalign.obs" class="el"> Observation directory name. |
| -suffix<em>FileSuffix</em><a | name="spr_vitalign.suffix" class="el"> File suffix of the observation files (without leading '.'). |
| -arcd<em>FileName</em><a | name="spr_vitalign.arcd" class="el"> Unit file name (.arcd or .cd format). |
| -h<em>FileName</em><a | name="spr_vitalign.h" class="el"> The input HMM file. |
| -g<em>FileName</em><a | name="spr_vitalign.g" class="el"> The input MVG file (gaussians). |
| -sel<em>FileName</em><a | name="spr_vitalign.sel" class="el"> The input select file name (tied gaussian). |
| -am_opt<em>Options</em><a | name="spr_vitalign.am_opt" class="el"> Extra options for loading the acoustic model. A non-default acoustic model can be selected by having '=<am_type>;' as first option. See cwr_am_tbl.c for a list of acoustic models available. |
| -top_n<em>Number(s)</em><a | name="spr_vitalign.top_n" class="el"> Only take the top-N gaussians into account when calculating output probabilities. If one value is given, it is used for all mixtures. Else a value per mixture must be given, separated by commas. Use '0' to set top_n to the number of gaussians in the mixture. |
| -rmg<em>rmg_params</em><a | name="spr_vitalign.rmg" class="el"> The parameters for the quick selection of gaussians. If one value is given, it is used for all mixtures. Else a value per mixture must be given, separated by commas. Use 'no' if no quick selection is wanted. See rm_gauss.c for a description of the parameters. |
| -LMout<em>Value</em><a | name="spr_vitalign.LMout" class="el"> Floor the state likelihoods of an observation using a fraction of the unconditional likelihood of the observation (weighted sum of the state likelihoods). Practically necessary if only few gaussians are evalutated (-top_n or -rmg options). The value given offset an automatically determined log10(fraction). Use -100 to turn the flooring off, and 0.0 to use the default.automatically. |
| -NOGSflag | no gauss sel Forgo the (sentence level) lexicon based Gaussian selection. The lexicon based Gaussian selection speeds up the decoding but may interfere with score normalization techniques that assume all Gaussians were evaluated. |
| -ci<em>CI-file</em><a | name="spr_vitalign.ci" class="el"> The unit description file. i.e. the lists the units (phones). Most program allow that both the CI-file and the CD-file (see the -cd option) are given as argument to the -ci option (seperated by a space). |
| -cd<em>CD-file</em><a | name="spr_vitalign.cd" class="el"> The file descibing the context dependencies (see also the -ci option). |
| -d<em>Dictionary</em><a | name="spr_vitalign.d" class="el"> Dictionary file name. |
| -unwind<em>unwind | format Define the parameters to modify the word transcriptions. See spr_cwr_lex_desc_read() for more details. |
Viterbi alignment
Plain Viterbi alignment of one observation (datafile) against a given word or unit sequence. Also an alignment of all entries in a corpus file against the correct model (spelling entry in the corpus file) can be done. The score (not for corpus files) and the path is written in text format to a segmentation file (directly usable as segmentation file for initseg, ...).
1.8.6