<HTML>
<BODY>
<PRE>

</PRE>
<H2>NAME</H2><PRE>
     lpcana  -  Performs  LPC  analysis  using  a  crude   pitch-
     synchronous method


</PRE>
<H2>SYNOPSIS</H2><PRE>
     <STRONG>lpcana</STRONG> [ <STRONG>-P</STRONG> <EM>param</EM>_<EM>file</EM> ] [ <STRONG>-r</STRONG> <EM>range</EM> ] [ <STRONG>-p</STRONG>  <STRONG>range</STRONG> ]  [   <STRONG>-</STRONG> <STRONG>m</STRONG>
     <EM>anal</EM>_<EM>method</EM> ] [ <STRONG>-F</STRONG> ] [ <STRONG>-x</STRONG> <EM>debug</EM>_<EM>level</EM> ] <EM>input</EM>_<EM>sd</EM> <EM>output</EM>_<EM>fea</EM>


</PRE>
<H2>DESCRIPTION</H2><PRE>
     Use of this program is not recommended.  For pitch  synchro-
     nous  analysis,  see  <STRONG><A HREF="../man1.html/epochs.1.html">epochs  (1 -ESPS)</A></STRONG> and <STRONG><A HREF="../man1.html/ps_ana.1.html">ps_ana (1-ESPS)</A></STRONG>.
     <EM>lpcana</EM> takes as input an ESPS sampled data <STRONG><A HREF="../man5.html/FEA_SD.5.html">FEA_SD (5 - ESPS)</A></STRONG>
     file and it computes LPC filter coefficients and the excita-
     tion parameters using a crude pitch-synchronous method.  The
     results  are stored in the <STRONG><A HREF="../man5.html/FEA_ANA.5.html">FEA_ANA (5-ESPS)</A></STRONG> file <EM>output</EM>_<EM>fea</EM>.
     For each analysis frame of data, <EM>lpcana</EM> generates reflection
     coefficients,  power,  and  voicing  information.  An output
     FEA_ANA record is generated for each pulse  (several  pulses
     per  analysis  frame),  with  enough  information  to enable
     <STRONG><A HREF="../man1.html/lpcsynt.1.html">lpcsynt (1-ESPS)</A></STRONG> to re-synthesize speech.  The nominal frame
     size  is  set  by the parameter <EM>lpc</EM>_<EM>frame</EM>_<EM>size</EM> but is deter-
     mined adaptively by the program.  If the <STRONG>-F</STRONG>  option  is  not
     used,  <EM>refcof</EM>  performs  an  additional spectral analysis on
     each pulse within the analysis frame and  puts  out  a  dif-
     ferent  set of reflection coefficients for each pulse (if <STRONG>-F</STRONG>
     is used, all pulses from the same analysis  frame  have  the
     same reflection coefficients).

     If "-" is used in place of <EM>input</EM>_<EM>sd</EM>, standard input is used.
     If  " - " is used in place of <EM>output</EM>_<EM>fea</EM>, standard output is
     used.


</PRE>
<H2>OPTIONS</H2><PRE>
     The following options are supported:

     <STRONG>-P</STRONG> <EM>param</EM>_<EM>file</EM>
          uses the parameter  file  <EM>param</EM>_<EM>file</EM>  rather  than  the
          default file <EM>params</EM>.

     <STRONG>-p</STRONG> <EM>range</EM>
          Selects a subrange of points to be analyzed. The  start
          and end points are defined with respect to the original
          SD file that is the source of the input  FEA_ANA  file.
          The  range  is  specified using the format <EM>start</EM>-<EM>end</EM> or
          <EM>start</EM>:<EM>end</EM> or <EM>start</EM>:+<EM>nan</EM>.  Either <EM>start</EM> or  <EM>end</EM>  may  be
          omitted,  in  which case the omitted parameter defaults
          respectively to the start or end of the input SD file.

     <STRONG>-r</STRONG> <EM>range</EM>
          <STRONG>r</STRONG> is a synonym for <STRONG>p</STRONG>.

     <STRONG>-m</STRONG> <EM>anal</EM>_<EM>method</EM>[mburg]
          Specifies the spectrum analysis method.  The default is
          the modified Burg method.  Also available are the auto-
          correlation method ("autoc"), ovariance method ("cov"),
          Burg   method   ("burg"),  fast  modified  Burg  method
          ("fburg"),   stuctured   covariance    ("strcov"    and
          "strcov1"),  and  vector Burg ("vburg", fast approxima-
          tion to structured covariance.  Of the  two  structured
          covariance methods, the first ("strcov" is considerably
          faster and  better  behaved;  "strcov1"  is  older  but
          included  as  it  may prove useful on occasion.  The <STRONG>-m</STRONG>
          option overrides the value that may be in the parameter
          file.  The default applies only if there is no value in
          the parameter file.

     <STRONG>-F</STRONG>    Specifies use of the framing method  of  the  previous
          version of <EM>refcof</EM>, in which the reflection coefficients
          for each pulse in an analysis frame are determined by a
          spectral  analysis  of  the entire frame.  If <STRONG>-F</STRONG> is not
          used, an additional spectrum analysis is performed  for
          each pulse.

     <STRONG>-x</STRONG> <EM>debug</EM>_<EM>level</EM> [0]
          option specifies that various information or  debugging
          messages be printed on standard error.


</PRE>
<H2>ESPS HEADER</H2><PRE>
     The generic header item <EM>start</EM>_<EM>time</EM> (type DOUBLE) is  written
     in the output file.  The value written is computed by taking
     the <EM>start</EM>_<EM>time</EM> value from the header of the input  file  (or
     zero,  if such a header item doesn't exist) and adding to it
     the offset time (from the beginning of the  input  file)  of
     the first point or record processed.  Unlike many other ESPS
     programs, <EM>lpcana</EM> does <EM>not</EM>  write  the  generic  header  item
     <EM>record</EM>_<EM>freq</EM>  (used by <EM>waves</EM>+ for time synchronization); this
     is because the output records  from  <EM>lpcana</EM>  result  from  a
     variable frame length analysis.

     The following generic header items are added (in addition to
     the standard ones (FEA_ANA-5):

          <EM>p</EM>_<EM>offset</EM> - <EM>integer</EM>

          How many points before the beginning of the  pulse  the
          spectrum analysis window is started.


          <EM>dcrem</EM> - <EM>string</EM>

          Yes means that the DC component was removed before  the
          data was analyzed.


          <EM>psynch</EM> - <EM>string</EM>

          Yes means that the spectrum analysis is done pitch syn-
          chronously.


          <EM>matsiz</EM> - <EM>integer</EM>

          The size of the autocorrelation matrix that was used in
          the spectrum analysis.



</PRE>
<H2>ESPS PARAMETERS</H2><PRE>
     The values of parameters obtained from  the  parameter  file
     are printed if the environment variable ESPS_VERBOSE is 3 or
     greater.  The default value is 3.  The following  parameters
     are read from the parameter file:

          <EM>start</EM> - <EM>integer</EM>

          This is the starting point  in  the  input  file.   Its
          value is superseded by a <STRONG>-p</STRONG> value. The default value is
          1.


          <EM>nan</EM> - <EM>integer</EM>

          This is the number of points to analyze. Its  value  is
          superseded by a <STRONG>-p</STRONG> value.


          <EM>lpc</EM>_<EM>filter</EM>_<EM>order</EM> - <EM>integer</EM>

          This is the order of the linear prediction filter  that
          represents the vocal tract.  The maximum allowable size
          is 20.  The default value is 10.

          <EM>lpc</EM>_<EM>frame</EM>_<EM>size</EM> - <EM>integer</EM>

          This is the nominal frame size for performing  spectral
          analysis. The default value is 160.

          <EM>minimum</EM>_<EM>pulse</EM>_<EM>length</EM> - <EM>integer</EM>

          This specifies the minimum pitch period. The  algorithm
          has  been well tested for the pitch range of 50-400 Hz.
          Thus a typical value of this variable is 20, assuming a
          sampling rate of 8000 Hz.  The default value is 20.

          <EM>method</EM> - <EM>string</EM>

          The spectrum analysis method  to  use.   The  available
          methods   are   autocorrelation  ("autoc"),  covariance
          ("covar"), Burg ("burg"), modified Burg ("mburg"), fast
          modified  Burg  method  ("fburg"), stuctured covariance
          ("strcov" and "strcov1"), and vector Burg  ("vburg",  a
          fast  approximation  to  structured  covariance.  If no
          value is given in the file, the modified Burg method is
          used  by  default.   The  <EM>method</EM>  is  not read from the
          parameter file if the command line option <STRONG>-m</STRONG> is used.


</PRE>
<H2>ESPS COMMON</H2><PRE>
     If the input is standard input, COMMON is not read. If  COM-
     MON  is  read and the command line input filename does match
     the filename listed in COMMON then, the following items  are
     read.  If  the  two  filenames do not match, then no further
     parameters are read from COMMON.

          <EM>start</EM> - <EM>integer</EM>

          This is the starting point in the input file.  Any  <STRONG>-</STRONG> <STRONG>p</STRONG>
          option value supersedes the COMMON specified value.

          <EM>nan</EM> - <EM>integer</EM>

          This is the number of points to analyze. A <STRONG>-p</STRONG> specified
          value supersedes the COMMON specified value.

     ESPS Common  processing  may  be  disabled  by  setting  the
     environment  variable USE_ESPS_COMMON to "off".  The default
     ESPS Common file is .espscom in the user's  home  directory.
     This  may  be  overidden by setting the environment variable
     ESPSCOM to the desired path.  User feedback of  Common  pro-
     cessing   is   determined   by   the   environment  variable
     ESPS_VERBOSE, with 0 causing no feedback and increasing lev-
     els causing increasingly detailed feedback.  If ESPS_VERBOSE
     is not defined, a default value of 3 is assumed.


</PRE>
<H2>COMMENTS</H2><PRE>
     The algorithm has been developed for speech sampled at  8000
     Hz.  It  works  best  when the data is recorded using a good
     microphone. Its quality is degraded, when working with tele-
     phone speech (especially if carbon microphone is used).

     The speech data is not pre-emphasized prior to LPC analysis,
     and  the  modified Burg method is used for computing the LPC
     parameters.


</PRE>
<H2>FUTURE CHANGES</H2><PRE>
     NONE.


</PRE>
<H2>BUGS</H2><PRE>
     If the input data consists of a long string  of  zeros,  the
     program bombs.


</PRE>
<H2>SEE ALSO</H2><PRE>
     <STRONG><A HREF="../man1.html/refcof.1.html">refcof (1-ESPS)</A></STRONG>, <STRONG><A HREF="../man1.html/lpcsynt.1.html">lpcsynt (1-ESPS)</A></STRONG>, <STRONG><A HREF="../man5.html/FEA_ANA.5.html">FEA_ANA (5-ESPS)</A></STRONG>,
     <STRONG><A HREF="../man5.html/FEA_SD.5.html">FEA_SD (5-ESPS)</A></STRONG>, <STRONG><A HREF="../man3.html/compute_rc.3.html">compute_rc (3-ESPSsp)</A></STRONG>


</PRE>
<H2>AUTHOR</H2><PRE>
     S. Shankar Narayan.  ESPS 3.0 modification by David  Burton.
     Modified for <STRONG><A HREF="../man3.html/compute_rc.3.html">compute_rc (3-ESPSsp)</A></STRONG> and for single-pulse out-
     put by John Shore.












































</PRE>
</BODY>
</HTML>
