man.diff

speech signal process tools

.\" Copyright (c) 1986-1990 Entropic Speech, Inc.
.\" Copyright (c) 1991 Entropic Research Laboratory, Inc.; All rights reserved
.\" @(#)addclass.1	1.3 4/2/97 ESI/ERL
.ds ]W (c) 1991 Entropic Research Laboratory, Inc.
.TH ADDCLASS 1\-ESPS 4/2/97
.ds ]W "\fI\s+4\ze\h'0.05'e\s-4\v'-0.4m'\fP\(*p\v'0.4m'\ Entropic Speech, Inc.
.ds ]Y "\fBERL\fP
.SH "NAME"
addclass - add a FEA_VQ codebook record to a FEA_VQ codebook file
.SH "SYNOPSIS"
.B addclass 
[
.BI \-x " debug_level"
] [
.BI \-p " field_rep"
] [
.BI \-f " field"
] [
.BI \-r " record_num"
] [
.BI \-s " source_name" 
] [
.BI \-t " source_type"
] [
.BI \-n " signal_name"
]
.I " infile.cbk"
.I " outfile.cbk"
.SH "DESCRIPTION"
.PP
\fIAddclass\fR adds the \fB\-r\fR specified codebook record from
\fIinfile.cbk\fR to \fIoutfile.cbk\fR. Note both \fIinfile.cbk\fR and
\fIoutfile.cbk\fR are FEA_VQ files.  If the output codebook file
doesn't exist, \fIaddclass\fR creates it.  If it does exist, in the
process of updating it, \fIaddclass\fP creates a temporary file is
created in the directory specified by the environment variable
ESPS_TEMP_PATH (default /usr/tmp).  \fIAddclass\fR also fills in five
fields in the output codebook record: \fIfield_rep, field,
source_name, source_type,\fR and \fIsignal_name\fR.  Any previous
contents of these fields is lost.
.PP
\fIField\fR contains the name of the field in the FEA file that
was used as source data to design the codebook.
\fIField_rep\fR is the \fIfield\fR's representation.
For example, the \fIspec_param\fR
field in FEA_ANA files has the following possible 
\fIfield_rep\fR representations:
AUTO, AFC, LSF, CEP, RC, and LAR.
\fISource_name\fR contains the name of the source for the data.
For speaker or speech recognition, this is the name of the speaker.
\fISource_type\fR contains a category descriptor of the source.
For example in speaker identification, \fIsource_type\fR could be either
male or female.
\fISignal_name\fR describes the type of signal that the source produced.
In speaker identification, for example, this is the identity of the
word that was spoken by the unknown speaker.
Note that all of these parameters may take on a value
of \fINONE\fR, and this is the default value.
.PP
\fIAddclass\fR also checks for equality among all the codebooks for
the following parameters:
\fIcodeword_dimen, cbk_struct, field,\fR and \fIfield_rep.\fR
If each of these fields is not the same in all codebooks, then
processing is stopped and an error message is printed.
.PP
If 
.I infile1.fea
is set equal to "\-", 
standard input is used for the input file. 
.SH OPTIONS
.PP
.TP
.BI \-x " debug_level" " \fR[0]\fP"
Values
greater than 0 cause messages to print to \fIstderr\fR.
.TP
.BI \-p " field_rep" " \fR[NONE]\fP"
A \fIfield\fR's representation may be specified by using this option.
.TP
.BI \-f " field" " \fR[NONE]\fP"
Specifies the field in the codebook's input data file that
the codebook represents.
.TP
.BI \-r " record_num" 
Specifies an integer record number within \fIinfile.cbk\fR
to copy to the output file. If no \fB\-r\fR value
is specified, the last record in the input codebook is copied.
.TP
.BI \-s " source_name" " \fR[NONE]\fP"
Specifies the source of the data that was used to design
the codebook.
.TP
.BI \-t " source_type" " \fR[NONE]\fP"
Specifies the source's category.
.TP
.BI \-n " signal_name" " \fR[NONE]\fP"
Specifies the type of signal that the source produced.
.SH "ESPS PARAMETERS"
.PP
The PARAMS file is not processed.
.SH ESPS COMMON
.PP
The Common file is not processed.
.SH ESPS HEADERS
.PP
\fIInfile.cbk\fR is added to the output header 
as a source.
.PP
See FEA_VQ(5\-ESPS) for a description
of the standard header items.
.SH "FUTURE CHANGES"
.PP
.SH WARNINGS
.PP
\fIAddclass\fR warns and exits if the input and output file type is not
correct.
.SH "SEE ALSO"
.PP
vqdes(1\-ESPS), vq(1\-ESPS), FEA_DST(5\-ESPS), FEA_VQ(5\-ESPS),
vqclassify(1\-ESPS), vqdst(1-ESPS)
.SH "BUGS"
.PP
None known.
.SH "AUTHOR"
.PP
Manual page by David Burton.
Program by Alan Parker.