stats.1t

speech signal process tools

.\" @(#)stats.1t	1.1 5/21/87 ESI
.TH STATS 1\-ESPS 5/12/87
.ds ]W "\fI\s+4\ze\h'0.05'e\s-4\v'-0.4m'\fP\(*p\v'0.4m'\ Entropic Speech, Inc.
.SH NAME
stats \- compute and print statistics from ESPS files
.SH SYNOPSIS
.B stats
[
.BI \-r " rec_range"
] [
.BI \-e " el_range"
] [
.BI \-m " magnitude"
] [
.B \-n
]
.I file
.SH DESCRIPTION
.PP
.I Stats
considers an ESPS file to be a series of vectors, without interpretation.
For sampled data files, the vector has only one element.
It computes the mean, standard deviation, maximum, and minimum value for
each element of the vector. It also determines the record number containing
the maximum and minimum values for each element. If \fIfile\fR is 
replaced by "\-", standard input is read. Here is a sample report:
.sp
.TS
c c c c c c c
c r c r c r r.
Element	Minimum	at	Maximum	at	Mean	Std.	Dev
1	-197	159	116	668	-6.76805	88.7037
2	1.42592	909	242139	112	11304.6	26212
3	-0.841206	432	0.991291	662	0.486449	0.375299
4	-0.804104	609	0.625042	631	-0.137957	0.328477
5	-0.580893	165	0.886137	898	0.230655	0.283118
6	-0.930659	182	0.698105	824	-0.00425471	0.294458
7	-0.586242	491	0.59219	400	-0.00573014	0.237584
8	-0.751755	397	0.575956	153	-0.129099	0.231132
9	-0.776561	149	0.510735	662	-0.107838	0.226078
10	-0.897471	394	0.337365	627	-0.21132	0.238978
11	-0.433536	219	0.583233	396	0.0971656	0.214986
12	-0.466768	722	0.474034	224	-0.0172287	0.142221
13	-0.369482	491	0.394765	674	0.0395599	0.141968
14	-0.459926	823	0.337641	874	-0.03448	0.119477
.TE
.sp
.PP
By default, all records in the file, and every element in each record,
are processed. The
.B \-r
option allows a specified range of records to be processed; the
.B \-e
option allows a specified range of elements to be processed.
As is usual
for ESPS programs, a range consists of two integers separated by colons,
a single integer (a range consisting of one value),
an integer followed by a colon (the number of records in the file, or the
number of elements in the vector, determines the end of the range),
or a colon followed by an integer (the default is to start the range at 1).
.PP
The \fB\-m\fR
option computes the average adjusted magnitude  (1/N  {sum from 1 to N} 
| x \- m |,  where m is the average value) for each element
in the specified element range. 
.PP
The
.B \-n
option suppresses the text header line; only the values are printed.
This may make it easier to use Unix tools to process the output.
.SH ESPS COMMON
The following items are read from the ESPS Common File:
.IP
.I "start - integer"
.IP
This is the starting record in the input file.
This value is not used if a range is specified on the command line.  If
the range is not specified on the command line and this parameter is
missing from ESPS Common then then the first record in the file is the
starting record.
.sp
.I "nan - integer"
.IP
This is the number of records (starting at start) to process.
This value is not used if a range is specified on the command line.  If
the range is not specified on the command line and this parameter is
missing from ESPS Common then then the 
entire file starting at "start" is processed.
.sp
.I "filename - string"
.IP
This is the name of the file to be processed if \fIfile\fP is omitted
from the command line.
.PP
The following items are written into the ESPS Common file:
.IP
.I "start - integer"
.IP
The starting record in the file that was processed.
.sp
.I "nan - integer"
.IP
The number of records in the selected range.
.sp
.I "prog - string"
.IP
This is the name of the program (\fIstats\fP in this case).
.sp
.I "filename - string"
.IP
This is the name of the file that was processed.
.SH DIAGNOSTICS
.PP
A fatal error occurs if the input file does not exist or is not an ESPS
file.
.SH AUTHOR
Joe Buck; Alan Parker; Dave Burton