nbest-pron-score.1

这是一款很好用的工具包

nbest-pron-score(1)                           nbest-pron-score(1)



NNAAMMEE
       nbest-pron-score  -  score pronunciations and pauses in N-
       best hypotheses

SSYYNNOOPPSSIISS
       nnbbeesstt--pprroonn--ssccoorree [--hheellpp] _o_p_t_i_o_n ...

DDEESSCCRRIIPPTTIIOONN
       nnbbeesstt--pprroonn--ssccoorree reads N-best lists and computes log prob-
       ability scores for the pronunciations and pauses contained
       in them.  Pronunciation scoring requires that  the  N-best
       lists  contain  phone  backtraces in "NBestList2.0" nnbbeesstt--
       ffoorrmmaatt(5).

       Pronunciation scores are computed from  the  probabilities
       in  a  dictionary.   Pauses  are  binned into three length
       classes (none, short, long) and scored according to a tri-
       gram  language  model  that conditions the pause length on
       the left and right neighboring words, in  that  order  (so
       that bigram backoff uses the left neighbor only).

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).

       --ttoolloowweerr
              Map all vocabulary to lowercase.   Useful  if  case
              conventions for text/counts and language model dif-
              fer.

       --mmuullttiiwwoorrddss
              Deal with N-best lists containing multiwords joined
              by  underscores.   This only affects pause scoring:
              if a word adjacent to a pause is a multiword and is
              not  in  the vocabulary of the pause LM, then it is
              split and only the component closest to  the  pause
              is conditioned on.

       --nnbbeesstt _f_i_l_e
              Score the N-best hypothese in _f_i_l_e.

       --rreessccoorree _f_i_l_e
              Same as --nnbbeesstt.

       --nnbbeesstt--ffiilleess _f_i_l_e
              Process all N-best list filenames listed in _f_i_l_e.

       --mmaaxx--nnbbeesstt _n
              Limits the number of hypotheses read from an N-best
              list.  Only the first _n hypotheses are processed.

       --ddiiccttiioonnaarryy _f_i_l_e
              Enable pronunciation scoring, using the  pronuncia-
              tion dictionary _f_i_l_e.  Each line contains a pronun-
              ciation in the format
                   _w_o_r_d [_p] _p_h_o_n_e ...
              The optional value _p is the pronunciation probabil-
              ity.  If the second field in a line is not a number
              the pronunciation is assumed  to  have  probability
              one.

       --iinnttllooggss
              Interpret  probabilities  in the dictionary as int-
              log-scaled log probabilities (as used  in  the  SRI
              Decipher(TM)  system),  rather than straight proba-
              bilities.

       --ppaauussee--llmm _f_i_l_e
              Enable pause scoring, using the pause LM in _f_i_l_e.

       --nnoo--ppaauussee _t_a_g
              The word used to represent the absence of  a  pause
              in the pause LM.

       --sshhoorrtt--ppaauussee _t_a_g
              The  word  used  to  represent a short pause in the
              pause LM.

       --lloonngg--ppaauussee _t_a_g
              The word used to represent  a  long  pause  in  the
              pause LM.

       --mmiinn--ppaauussee--dduurr _T
              The  minimum duration, in seconds, for a non-speech
              region to be considered a (short) pause.

       --lloonngg--ppaauussee--dduurr _T
              The duration, in second, above which  a  non-speech
              region is considered a "long" pause.

       The  default values for pause tags and duration thresholds
       are printed by the --hheellpp option.

       --pprroonn--ssccoorree--ddiirr _d_i_r
              Write pronunciation scores to _d_i_r  when  processing
              multiple   N-best  lists,  using  output  filenames
              derived from the input files.

       --ppaauussee--ssccoorree--ddiirr _d_i_r
              Write pause scores to _d_i_r when processing  multiple
              N-best  lists,  using output filenames derived from
              the input files.

       --ppaauussee--ssccoorree--wweeiigghhtt _W
              Add pause LM scores  to  the  pronunciation  scores
              after multiplying them by _W.  This creates a single
              weighted combination of both models.  Pause  scores
              can   still  be  output  separately  by  specifying
              --ppaauussee--ssccoorree--ddiirr.

SSEEEE AALLSSOO
       nbest-format(5),   nbest-scripts(1),    nbest-optimize(1),
       ngram(1).
       D.  Vergyri,  A.  Stolcke, V. R. R. Gadde, L. Ferrer, & E.
       Shriberg,  ``Prosodic  Knowledge  Sources  for   Automatic
       Speech  Recognition''.   _P_r_o_c_.  _I_E_E_E _I_n_t_l_. _C_o_n_f_. _o_n _A_c_o_u_s_-
       _t_i_c_s_, _S_p_e_e_c_h _a_n_d _S_i_g_n_a_l _P_r_o_c_e_s_s_i_n_g, Hong Kong, April 2003.

BBUUGGSS
       The  binning of pause lengths into three classes should be
       generalized.

AAUUTTHHOORR
       Andreas Stolcke <[e-mail]>.
       Copyright 2002-2004 SRI International



SRILM Tools        $Date: 2004/12/03 17:59:01 $nbest-pron-score(1)