nbest-lattice.1

这是一款很好用的工具包

nbest-lattice(1)                                 nbest-lattice(1)



NNAAMMEE
       nbest-lattice - rescore N-best lists and lattices

SSYYNNOOPPSSIISS
       nnbbeesstt--llaattttiiccee [--hheellpp] option ...

DDEESSCCRRIIPPTTIIOONN
       nnbbeesstt--llaattttiiccee  rescores  N-best  lists  or optimizes word-
       level recognition scores  (as  opposed  to  sentence-level
       scores).   There  are two rescoring modes.  In _N_-_b_e_s_t _w_o_r_d
       _e_r_r_o_r _m_i_n_i_m_i_z_a_t_i_o_n mode, the program computes  the  poste-
       rior  expected  word error for each hypothesis relative to
       all hypotheses in the N-best list, choosing the  one  with
       the lowest value.

       In  _l_a_t_t_i_c_e _w_o_r_d _e_r_r_o_r _m_i_n_i_m_i_z_a_t_i_o_n mode, the program con-
       structs a word lattice from all the N-best hypotheses  and
       extracts  the  path  with  the lowest expected word error.
       This is similar to  N-best  word  error  minimization  but
       allows  hypotheses  not  contained  in the N-best list.  A
       variant of this mode uses a word  ``mesh''  instead  of  a
       word  lattice,  in which all hypotheses are aligned into a
       grid of word positions, and one is allowed to chose a word
       from  each  grid  position,  thus allowing an even greater
       number of potential hypotheses.

OOPPTTIIOONNSS
       Each filename argument can be an ASCII  file,  or  a  com-
       pressed file (name ending in .Z or .gz), or ``-'' to indi-
       cate stdin/stdout.

       --hheellpp  Print option summary.

       --vveerrssiioonn
              Print version information.

       --ddeebbuugg _l_e_v_e_l
              Controls the  amount  of  output  (the  higher  the
              _l_e_v_e_l,  the  more).   At level 1, the expected word
              error counts for the chosen hypotheses are printed.
              At  level  2,  the word posterior probabilities are
              printed in addition (only for lattice mode, similar
              to --dduummpp--ppoosstteerriioorrss).

       --wweerr   Chooses N-best word error minimization mode.

       --llaattttiiccee--wweerr
              Chooses  lattice  word error minimization mode (the
              default).

       --uussee--mmeesshh
              Choose the variant of lattice mode that  uses  word
              meshes instead of simple lattices.

       --ddeelleettiioonn--bbiiaass _D
              Causes  the probabilities of deletions to be biased
              by a factor _D in doing mesh-based word error  mini-
              mization.   This  controls  the  trade-off  between
              insertion and deletion errors.  The  default  is  1
              (no bias).

       --rreessccoorree _f_i_l_e
              Reads  the  N-best list from _f_i_l_e.  The N-best list
              can be in any of the formats  described  in  nnbbeesstt--
              ffoorrmmaatt(5).

       --nnbbeesstt _f_i_l_e
              A synonym for --rreessccoorree.

       --wwrriittee--nnbbeesstt _f_i_l_e
              Outputs  the  N-best  list to a file, after sorting
              and processing (for validation or format conversion
              purposes).

       --nnbbeesstt--ffiilleess _f_i_l_e_-_l_i_s_t
              Rescores  multiple N-best lists whose filenames are
              read from _f_i_l_e_-_l_i_s_t.

       --wwrriittee--nnbbeesstt--ddiirr _d_i_r_e_c_t_o_r_y
              Outputs N-best lists to _d_i_r_e_c_t_o_r_y, to  files  named
              after  the input N-best lists, for when multiple N-
              best lists are processed (see --nnbbeesstt--ffiilleess).

       --wwrriittee--vvooccaabb _f_i_l_e
              Outputs vocabulary used in N-best list.

       --ddeecciipphheerr--nnbbeesstt
              Output N-best  list  in  Decipher  nnbbeesstt--ffoorrmmaatt(5),
              rather  than the default native SRILM format.  (All
              N-best formats are accepted for input regardless of
              this option.)

       --nnoo--rreessccoorree
              Suppress  rescoring of lattices; useful if only the
              operations of lattice/N-best  list  reading/writing
              are desired.

       --mmaaxx--nnbbeesstt _n
              Limits  the  number of hypotheses read from each N-
              best list to the first _n.

       --mmaaxx--rreessccoorree _m
              In  N-best  mode,  only  choose  among  the  top  _m
              hypotheses  when  optimizing  word  error.  This is
              convenient to limit  computation  for  long  N-best
              lists.   The  cutoff  is  made  after  reading  all
              hypotheses (subject to --mmaaxx--nnbbeesstt)  and  reordering
              them according to the posterior probabilities.
              The worst-case time taken in N-best error minimiza-
              tion is proportional to _m times _n, where _n  is  the
              length  of  the  N-best list (or the value given to
              --mmaaxx--nnbbeesstt).  However, in practice the average time
              per sentence is independent of _m, so this option is
              usually not necessary.
              In lattice mode,  only  align  the  top  _m  scoring
              hypotheses (after reweighting and sorting) into the
              lattice.

       --ppoosstteerriioorr--pprruunnee _t_h_r_e_s_h_o_l_d
              Don't process N-best  hypotheses  whose  cumulative
              posterior  probability is below _t_h_r_e_s_h_o_l_d.  This is
              another strategy to speed up the algorithm.

       --nnoo--rreeoorrddeerr
              Process N-best hypotheses in  the  order  in  which
              they  appear.   By  default,  hypotheses  are first
              sorted by their aggregate scores.

       --nnbbeesstt--bbaacckkttrraaccee
              Preserve    backtrace    information    (word-level
              timemarks  and  scores)  when  reading N-best lists
              containing such information (see  nnbbeesstt--ffoorrmmaatt(5)).
              The  default is to ignore backtrace information and
              record only  sentence-level  scores  and  the  word
              identities.

       --oouuttppuutt--ccttmm
              Output  word  hypotheses  in NIST CTM (conversation
              time mark) format.  Note that word start times will
              be  relative  to the segment start times, the first
              column will contain the N-best  filename,  and  the
              channel  field  is  always  1.  The word confidence
              field  contains  posterior   probabilities.    This
              option also implies --nnbbeesstt--bbaacckkttrraaccee.

       --rreessccoorree--llmmww _l_m_w
              Sets  the  language  model weight used in combining
              the language model log probabilities with  acoustic
              log probabilities (only relevant if separate scores
              are given in the N-best input).

       --rreessccoorree--wwttww _w_t_w
              Sets the word transition weight used to weight  the
              number of words relative to the acoustic log proba-
              bilities (only  relevant  if  separate  scores  are
              given in the N-best input).
              If  --nnoo--rreeoorrddeerr is not specified, and either _l_m_w or
              _w_t_w are specified to  be  non-zero,  the  aggregate
              scores  are  recomputed using those weights; other-
              wise aggregate scores supplied in the input  N-best
              lists are used to sort hypotheses.

       --ppoosstteerriioorr--ssccaallee _s_c_a_l_e
              Divide  the  total weighted log score by _s_c_a_l_e when
              computing normalized posterior probabilities.  This
              controls  the peakedness of the posterior distribu-
              tion.  The default value is whatever was chosen for
              --rreessccoorree--llmmww,  so  that  language  model scores are
              scaled to have weight 1, and acoustic  scores  have
              weight 1/_l_m_w.

       --ppoosstteerriioorr--aammww _a_m_w
              Sets the acoustic model weight for computing poste-
              riors; the default is 1.  This  and  the  next  two
              options  allow  posteriors  to  be computed using a
              different weighting than that used in  ranking  and
              reordering the hypotheses.

       --ppoosstteerriioorr--llmmww _l_m_w
              Sets the language model weight for computing poste-
              riors.  The default is to use whatever  was  speci-
              fied for --rreessccoorree--llmmww.

       --ppoosstteerriioorr--wwttww _w_t_w
              Sets  the word transition weight for computing pos-
              teriors.  The default is to use whatever was speci-
              fied for --rreessccoorree--wwttww.