HTK-HVite

HTK-HVite

This operation is similar to the HLEd word-to-phone mapping operation , however in this case the HVite command can consider all pronunciations for each word (in the case where a word has more than one pronunciations), and then output the pronunciation that best matches the acoustic data.

HVite的目的在于为每一个word后加一个sp
比如：
"*/sample1.lab"
DIAL
ONE
TWO
THREE
FOUR
FIVE
SIX
SEVEN
EIGHT
NINE
OH
ZERO
.
经过word-to-phone转换后：
"*/sample1.lab"
sil
d
ay
l
w
ah
n
t
uw
th
r
iy
f
ow
r
f
ay
v
s
ih
k
s
s
eh
v
ih
n
ey
t
n
ay
n
ow
z
iy
r
ow
sil
.
每个词之间有一个小的sp，处理后的结果为：
"*/sample1.lab"
sil
d
ay
l
sp
w
ah
n
sp
t
uw
sp
th
r
iy
sp
f
ow
r
sp
f
ay
v
sp
s
ih
k
s
sp
s
eh
v
ih
n
sp
ey
t
sp
n
ay
n
sp
ow
sp
z
iy
r
ow
sp
sil
.

指令：
$ HVite -A -D -T 1 -l '*' -o SWT -b SENT-END -C config -H hmm7/macros -H hmm7/h
mmdefs -i aligned.mlf -m -t 250.0 150.0 1000.0 -y lab -a -I words.mlf -S train.
scp dict monophones1>HVite_log
-l dir :
   This specifies the directory to store the output label files.If this option is not used then HVite will store the label files in the same directory as the data.In particular,setting the option -l ’*’ will cause a label file named xxx to be prefixed by the patternn "*/xxx" in the output MLF file. This is useful for generating MLFs which are independent of the location of the corresponding data files.

-o s :
     -o s    output label formating NCSTWMX
Choose how the output labels should be formatted. s is a string with certain letters (from NSCTWM) indicating binary flags that control formatting options.
  N normalise acoustic scores by dividing by the duration (in frames) of the segment.
  S remove scores from output label. By default scores will be set to the total likelihood of the segment.
  C Set the transcription labels to start and end on frame centres. By default start times are set to the start time of the frame and end times are set to the end time of the frame.
  T Do not include times in output label files. W Do not include words in output label files when performing state or model alignment.  
  M Do not include model names in output label files when performing state
and model alignment.

   -b s    def s as utterance boundary word 

   -H mmf  Load HMM macro file mmf

   -i s    Output transcriptions to MLF s 

   -y s    output label file extension

   -a      align from label files               off


-b s    def s as utterance boundary word     none
-c f    tied mixture pruning threshold       10.0
-d s    dir to find hmm definitions          current
-e      save direct audio rec output         off
-f      output full state alignment          off
-g      enable audio replay                  off
-h s    set speaker name pattern             *.mfc
-i s    Output transcriptions to MLF s       off
-j i    Online MLLR adaptation               off
         Perform update every i utterances
-k      use an input transform               off
-l s    dir to store label/lattice files     current
-m      output model alignment               off
-n i [N] N-best recognition (using i tokens) off
-o s    output label formating NCSTWMX       none
-p f    inter model trans penalty (log)      0.0
-q s    output lattice formating ABtvaldmn   tvaldmn
-r f    pronunciation prob scale factor      1.0
-s f    grammar scale factor                 1.0
-t f [f f] set pruning threshold             0.0
-u i    set pruning max active               0
-v f    set word end pruning threshold       0.0
-w [s]  recognise from network               off
-x s    extension for hmm files              none
-y s    output label file extension          rec
-z s    generate lattices with extension s   off
-A      Print command line arguments         off
-B      Save HMMs/transforms as binary       off
-C cf   Set config file to cf                default
-D      Display configuration variables      off
-E s [s] set dir for parent xform to s       off
         and optional extension
-F fmt  Set source data format to fmt        as config
-G fmt  Set source label format to fmt       as config
-H mmf  Load HMM macro file mmf
-I mlf  Load master label file mlf
-J s [s] set dir for input xform to s        none
         and optional extension
-K s [s] set dir for output xform to s       none
         and optional extension
-L dir  Set input label (or net) dir         current
-P      Set target label format to fmt       as config
-S f    Set script file to f                 none
-T N    Set trace flags to N                 0
-V      Print version information            off
-X ext  Set input label (or net) file ext    lab