intro.doc

speech signal process tools

                         ENTROPIC RESEARCH LABORATORY, INC.






                                INTRODUCTION TO THE

                      ENTROPIC SIGNAL PROCESSING SYSTEM (ESPS)


                            John Shore  and Alan Parker


                         Entropic Research Laboratory, Inc.
                           Washington Research Laboratory
                       600 Pennsylvania Ave. S.E., Suite 202
                               Washington, D.C. 20003
                                   [phone]




       1 .  INTRODUCTION

       The Entropic Signal Processing System (ESPS) is a set  of  interacting
       signal  processing programs.  It is a third-generation system that was
       developed over a three-year period with these main goals:

         +  to provide a useful set of easy-to-use programs for  direct  sup-
            port  of  such  traditional  operations as analog I/O, filtering,
            spectrum analysis, quantization, pattern  classification,  signal
            editing, file editing, plotting, etc;

         +  to provide a toolkit that supports the rapid development  of  new
            signal processing programs;

         +  to exploit UNIXr| tools and features (e.g., pipes)

         +  to facilitate record keeping in such a way that ESPS  files  con-
            tain a complete history of the origin of the data in the file and
            of all intermediate processing steps;

         +  to allow easy user-extensions to ESPS file types  (including  the
            creation of new file types) in such a way that the extensions are
            supported automatically  by  standard  ESPS  plotting,  printing,
            editing, and database programs;
       ____________________
          c Copyright  1987-1993  Entropic  Research  Laboratory,  Inc.;  All
       rights reserved.












       INTRODUCTION TO ESPS                                            page 2


       ESPS currently includes approximately 85 user-level  programs  (called
       from  the  UNIX  shell) for file manipulation, sampled-data I/O, plot-
       ting, and signal processing operations.  Also included is a library of
       approximately  200 functions that support the modification of existing
       programs and the development of new programs.  The ESPS  library  con-
       tains  functions  that perform signal processing primitives as well as
       functions that support the creation and manipulation  of  ESPS  files.
       ESPS  provides  5 built-in file types; it also provides the ability to
       modify these file types and to create new file types.

       The ESPS library was developed with disciplined  software  engineering
       methods,  including information-hiding.  The library functions perform
       extensive error-checking, and a complete LINT library is included.

       ESPS main programs and most of the library functions are written in C.
       However,  the  ESPS  library  also supports FORTRAN so that new signal
       processing operations can be written either in C or FORTRAN.

       2 .  ESPS DOCUMENTATION

       The principal documentation for ESPS is the  ESPS  Users  Manual.  [1]
       This  contains UNIX-style manual pages for all programs, library func-
       tions, and file types.  The ESPS Manual is organized in the usual UNIX
       style,  with  Section  1  describing  user-level  programs,  Section 3
       describing the ESPS library, and Section 5 describing ESPS  file  for-
       mats.   Section 3 is further divided into two sections, one for signal
       processing functions and one for ESPS utility  functions.   References
       to  ESPS programs, library functions, and file types usually include a
       parenthetical pointer to the section of the ESPS Manual in  which  the
       manual  page  is located.  For example, "(3-ESPS)" indicates Section 3
       of the ESPS Manual.

       The entire ESPS Manual is available on-line and can be viewed with the
       eman  (1-ESPS) program, which operates just like the UNIX man (1) pro-
       gram (including the -k option to search for  programs  whose  descrip-
       tions contain a given keyword).

       The ESPS documentation also includes a series of  technical  memoranda
       that explain ESPS data files, [2] ESPS parameter and common files, [3]
       and ESPS feature files,  [4] files.  Also provided are  ESPS  program-
       ming  guidelines,  [5]  an  applications note on filtering, [6] and an
       applications note on ESPS record keeping. [7]

       3 .  ESPS FILES

       ESPS files have a common structure  consisting  of  a  variable-length
       header  followed  by  data  records.   Within  a single ESPS file, the
       records have a fixed-length that depends on information in the header.
       Record-lengths vary, however, even among files of the same basic type.
       ____________________
          | UNIX is a trademark of AT&T Bell Laboratories




       Version 3.6                      ERL                           1/22/93







       INTRODUCTION TO ESPS                                            page 3


       The ESPS header is the primary means by which record-keeping  is  per-
       formed  within ESPS.  An ESPS header includes the values of all impor-
       tant parameters to the program that created the ESPS  file.   It  also
       includes the headers of any files that were inputs to the program that
       created the file.  Since the header is a  recursive  structure,  these
       included headers themselves may contain the headers of previous source
       files.  It follows that most ESPS files contain a complete history  of
       the origin of the data in the file and all the intermediate processing
       steps.  In addition to this formal record-keeping mechanism,  informal
       comments  can  be  added  to  any ESPS file header by means of comment
       (1-ESPS).  For details about  the  ESPS  history  mechanism,  see  the
       applications note. [7]

       ESPS file headers and data records are viewed in ASCII form  by  means
       of  the  program  psps  (1-ESPS),  which  includes options for viewing
       recursively all of the embedded headers.  The output from psps identi-
       fies  values  from header and data-records fields by name. Within pro-
       grams, the fields are also referred to by name.

       On some systems, ESPS files can be created directly from analog inputs
       and  played  out  over digital to analog converters.  Also, conversion
       programs are available for ASCII files, ILSr| sampled-data files,  and
       MASSCOMPr Laboratory Workbench (LWB)| files.

       3 .1 .  Built-in File Types

       Older versions of ESPS contained the following built-in file types:

         SD      contains sampled data

         FILT    contains digital filters

         SCBK    contains scalar quantization codebooks

         FEA     contains arbitrary features

       Except for FEA files, which are discussed further below, these  built-
       in  files  have  a  predefined  record structure.  That is, their data
       records comprise fields with pre-defined names and  data  types.   The
       definitions  are suitable for the types of data stored in these built-
       in file types.  Similarly, the headers of the built-in  types  contain
       pre-defined fields for relevant information.  However, any program can
       add additional header fields through the use of generic  header  items
       (see  Section 7.1.3).  Generic header items can also be added by means
       of the user-level program addgen (1-ESPS).


       ____________________
          | ILS is a trademark of Signal Technology, Inc.
          |  MASSCOMP  and  LWB  are  trademarks  of  Massachusetts  Computer
       Corporation.




       Version 3.6                      ERL                           1/22/93







       INTRODUCTION TO ESPS                                            page 4


       Section 5 of the ESPS Manual contains detailed information about  what
       is stored in the various ESPS file types.

       3 .2 .  Feature Files and User-Defined File Types

       The current version of ESPS uses feature files for all data.  The  FEA
       file  type  is  a  general purpose file type that can be used to store
       arbitrary information in fixed-length records; optionally, the records
       can  point  to positions in various source files.  Unlike the case for
       the other ESPS file types - the names, sizes, and data  types  of  the
       fields  in  FEA  records  are  not  pre-defined.  Because of this, and
       because of the ways in which FEA files are supported by standard  ESPS
       file  manipulation  and  plotting  programs,  FEA files can be used to
       create user-defined ESPS files that have all of the advantages of  the
       built-in  file  types.   This  can  be  done by using the ESPS library
       whether or not ESPS source is available. [4]

       We call FEA files self-describing because the structure of a FEA  file
       is defined in the file header, rather than in some external specifica-
       tion.  In particular, the self-description portion of the  FEA  header
       is  contained  in  the  type-specific  section  of  the header, and is
       displayed by fea_element (1-ESPS), or  by  psps  (1-ESPS)  withthe  -v
       option.  In many cases, non-ESPS binary files can be converted to ESPS
       FEA files using addfeahd (1-ESPS).

       When a new file type is implemented using FEA files, it is referred to
       as  a  FEA  file  sub-type.  Several such implementations are provided
       with ESPS, including the following:

         FEA_SD
              used for single and multi-channel sampled data (including  com-
              plex sampled data).

         FEA_SPEC
              used for storing spectral information, such as  power  spectra.
              Each  record  contains  a  vector of spectral values (powers or
              power densities in the case of power  spectra),  together  with
              certain  optional  fields - total power, an imaginary-part vec-
              tor, number of frequencies, and a vector of frequencies -  that
              are present or absent according to the values of certain header
              items;

         FEA_ANA
              suitable for speech processing;  contains  spectral  parameters
              (various  representations),  pulse lengths, powers, and voicing
              information;

         FEA_VQ
              contains vector quantization codebooks;

         FEA_FILT
              contains digital filters;



       Version 3.6                      ERL                           1/22/93







       INTRODUCTION TO ESPS                                            page 5


         FEA_STAT
              contains statistical  information  (means,  covariances,  etc.)
              suitable for pattern-classification;

       By using access functions provided with the ESPS  Library,  users  can
       implement their own FEA file sub-types.  They can also extend existing
       FEA file types, like those mentioned above, by adding  whatever  addi-
       tional record fields are desired.

       The power of FEA files is enhanced by the existence of  ESPS  programs
       like the following, which work on any FEA file (including user-defined
       or user-extended FEA files):

         psps(1-ESPS)
              produces ASCII  representation  of  any  ESPS  file  (including
              headers), with all fields referred to by name;

         genplot(1-ESPS)
              plots information from FEA file records in precise,  multi-line
              format;

         fea_edit(1-ESPS)
              permits you to edit an ASCII representation of a FEA file using
              the  editor of your choice; Thus, you can modify the data in an
              existing FEA file or create a new  file  with  arbitrary  data.
              Both file headers and data records can be edited with fea_edit.

         select(1-ESPS)
              treats ESPS FEA files as  a  database  and  allows  interactive
              selection   of  records  that  satisfy  user-supplied  queries;
              selected records can be output to another FEA  file;  Here  are
              some examples of queries on an input FEA_ANA file:

              ->select (spec_param[0] > .96) && (frame_type == "VOICED")
                      123 records selected

              ->select ((spec_param[0] + spec_param[1]) > 1.5) && (frame_len > 100)
                      21 records selected

              ->select (spec_param[0] > spec_param[1]) && (log(sqrt(raw_power[0])) > 7)
                      13 records selected

              ->select (mean(raw_power) > 1.75E10) || (raw_power[0] > 1.75E10)
                      100 records selected

              Here, "spec_param", "frame_type", "frame_len", and  "raw_power"
              are the names of fields in a FEA_ANA file.  Similar queries can
              be constructed for any FEA file using the names  of  fields  in
              the  FEA  file.   Queries can be constructed using simple rela-
              tional expressions  involving  the  elements  in  FEA  records,
              built-in  arithmetic  operators (*,+,-,/,^), built in functions
              (sin, cos, tan, atan, log, log10, exp, sqrt), and  user-defined
              functions  implemented  as external programs.  Also, the "eval"


       Version 3.6                      ERL                           1/22/93







       INTRODUCTION TO ESPS                                            page 6


              command in select allows you to  specify  arbitrary  arithmetic
              expressions  on  the FEA fields and have the results printed on
              standard output.

         fea_stat(1-ESPS)
              Computes means,  covariances,  eigenvalues,  eigenvectors,  and
              other  statistics  from  the  records  in a FEA file; outputs a
              FEA_STAT file;

         classify(1-ESPS)
              performs classification of FEA file records  using  information
              in FEA_STAT files;

       Another useful aspect of FEA files  is  a  mechanism  called  "segment
       labelling",  which  makes  it  easy to associate FEA file records with
       positions or ranges in sampled-data files.  If a FEA file is  declared
       to  be  segment-labelled,  then its records automatically include file
       name, starting position, and length fields (these need to be filled in
       by  user  programs,  but  the field support is there).  This mechanism
       could be used, for example, to support  a  database  of  phonetically-
       labelled  sampled  data  files by implementing a FEA file containing a
       phoneme code  (or  string)  in  each  segment-labelled  record.   Then
       select(1-ESPS)  could  be  used  to  process sets of these files.  For
       example, the select command

               ->select phoneme == "AW" || phoneme == "OW"

       would yield records that provide file names, starting  positions,  and
       lengths for all occurrences of the two named phonemes.

       4 .  ESPS PROGRAMS AND UNIX PIPES

       So that pipes can be used, most ESPS programs are written so that they
       use  standard input and output as well as specific files.  As an exam-
       ple, consider the following sequence of ESPS commands:

               %refcof data.sd data.fana
               %me_spec data.fana data.spec
               %plotspec data.spec


       The first command causes the  reflection  coefficients  computed  from
       data  frames  in  the SD file data.sd to be placed in the FEA_ANA file
       data.fana.  The second  command  computes  spectra  for  each  set  of
       reflection  coefficients  and  puts  the  results  in  the  SPEC  file
       data.spec.  The third command plots all  the  spectra  on  the  screen
       using  a single set of axes.  If there is no need to retain the inter-
       mediate FEA_ANA  and  FEA_SPEC  files,  one  could  do  the  following
       instead:

               %refcof data.sd - | me_spec - - | plotspec -




       Version 3.6                      ERL                           1/22/93