Edinburgh Speech Tools
2.4-release
All
Classes
Functions
Variables
Typedefs
Enumerations
Enumerator
Friends
Pages
spectgen_main.cc
1
/*************************************************************************/
2
/* */
3
/* Centre for Speech Technology Research */
4
/* University of Edinburgh, UK */
5
/* Copyright (c) 1995,1996 */
6
/* All Rights Reserved. */
7
/* */
8
/* Permission is hereby granted, free of charge, to use and distribute */
9
/* this software and its documentation without restriction, including */
10
/* without limitation the rights to use, copy, modify, merge, publish, */
11
/* distribute, sublicense, and/or sell copies of this work, and to */
12
/* permit persons to whom this work is furnished to do so, subject to */
13
/* the following conditions: */
14
/* 1. The code must retain the above copyright notice, this list of */
15
/* conditions and the following disclaimer. */
16
/* 2. Any modifications must be clearly marked as such. */
17
/* 3. Original authors' names are not deleted. */
18
/* 4. The authors' names are not used to endorse or promote products */
19
/* derived from this software without specific prior written */
20
/* permission. */
21
/* */
22
/* THE UNIVERSITY OF EDINBURGH AND THE CONTRIBUTORS TO THIS WORK */
23
/* DISCLAIM ALL WARRANTIES WITH REGARD TO THIS SOFTWARE, INCLUDING */
24
/* ALL IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS, IN NO EVENT */
25
/* SHALL THE UNIVERSITY OF EDINBURGH NOR THE CONTRIBUTORS BE LIABLE */
26
/* FOR ANY SPECIAL, INDIRECT OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES */
27
/* WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN */
28
/* AN ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, */
29
/* ARISING OUT OF OR IN CONNECTION WITH THE USE OR PERFORMANCE OF */
30
/* THIS SOFTWARE. */
31
/* */
32
/*************************************************************************/
33
/* Author: Paul Taylor */
34
/* Date : April 1995 */
35
/*-----------------------------------------------------------------------*/
36
/* Generate feature vectors */
37
/* */
38
/*=======================================================================*/
39
40
#include "EST.h"
41
#include "EST_cmd_line_options.h"
42
#include "sigpr/EST_spectrogram.h"
43
44
#define DEFAULT_FRAME_SIZE 0.001
45
#define DEFAULT_FRAME_LENGTH 0.008
46
#define DEFAULT_ORDER 256
47
#define DEFAULT_PREEMPH 0.94
48
49
void
set_options(
EST_Features
&op,
EST_Option
&al);
50
51
/** @name <command>spectgen</command> <emphasis>Make spectrograms</emphasis>
52
* @id spectgen-manual
53
* @toc
54
*/
55
56
//@{
57
58
/**@name Synopsis
59
*/
60
//@{
61
62
//@synopsis
63
64
/**
65
spectgen is used to create spectrograms, which are 3d plots of
66
amplitude against time and frequency. Spectgen takes a waveform and
67
produces a track, where each channel represents one frequency bin.
68
69
By default spectgen produces a "wide-band" spectrogram, that is one
70
with high time resolution and low frequency resolution. "Narrow-band"
71
spectrograms can be produced by using the -shift and -length options.
72
73
Typical values for -shift and -length are:
74
75
76
77
*/
78
79
//@}
80
81
/**@name Options
82
*/
83
//@{
84
85
//@options
86
87
//@}
88
89
90
int
main(
int
argc,
char
*argv[])
91
{
92
EST_String
out_file;
93
EST_StrList
files;
94
EST_Option
al;
95
EST_Features
op;
96
int
orig_sr;
97
98
EST_Wave
sig;
99
EST_Track
spec;
100
101
parse_command_line
102
(argc, argv,
103
EST_String
(
"[input file] -o [output file]\n"
)+
104
"Summary: make spectrogram\n"
+
105
"use \"-\" to make input and output files stdin/out\n"
+
106
"-h Options help\n"
+
107
options_wave_input()+
108
"\n"
+
109
options_track_output()+
110
"-shift <float> frame spacing in seconds for fixed frame analysis. This \n"
111
" doesn't have to be the same as the output file spacing - the \n"
112
" S option can be used to resample the track before saving \n"
113
" default: "
+ftoString(DEFAULT_FRAME_SIZE) +
"\n\n"
114
"-length <float> input frame length in milliseconds\n"
+
115
"-sr <float> range in which output values should lie\n"
+
116
"-slow slow FFT code\n"
+
117
"-w <float> white cut off (0.0 to 1.0)\n"
+
118
"-b <float> black cut off (0.0 to 1.0)\n"
+
119
"-raw Don't perform any scaling\n"
+
120
"-order <int> cepstral order\n"
, files, al);
121
122
out_file = al.
present
(
"-o"
) ? al.
val
(
"-o"
) : (
EST_String
)
"-"
;
123
set_options(op, al);
124
125
if
(read_wave(sig, files.
first
(), al) != format_ok)
126
exit(-1);
127
orig_sr = sig.
sample_rate
();
128
129
make_spectrogram(sig, spec, op);
130
131
spec.
save
(out_file, al.
val
(
"-otype"
, 0));
132
133
return
0;
134
}
135
136
void
set_options(
EST_Features
&op,
EST_Option
&al)
137
{
138
op.
set
(
"frame_shift"
, DEFAULT_FRAME_SIZE);
139
op.
set
(
"frame_length"
, DEFAULT_FRAME_LENGTH);
140
op.
set
(
"preemph"
, DEFAULT_PREEMPH);
141
op.
set
(
"frame_order"
, DEFAULT_ORDER);
142
143
if
(al.
present
(
"-shift"
))
144
op.
set
(
"frame_shift"
, al.
fval
(
"-shift"
));
145
146
if
(al.
present
(
"-length"
))
147
op.
set
(
"frame_length"
, al.
fval
(
"-length"
));
148
149
if
(al.
present
(
"-order"
))
150
op.
set
(
"frame_order"
, al.
fval
(
"-order"
));
151
152
if
(al.
present
(
"-sr"
))
153
op.
set
(
"sp_range"
, al.
fval
(
"-sr"
));
154
155
if
(al.
present
(
"-w"
))
156
op.
set
(
"sp_wcut"
, al.
fval
(
"-w"
));
157
158
if
(al.
present
(
"-b"
))
159
op.
set
(
"sp_bcut"
, al.
fval
(
"-b"
));
160
161
if
(al.
present
(
"-preemph"
))
162
op.
set
(
"preemph"
, al.
fval
(
"-preemph"
, 1));
163
164
if
(al.
present
(
"-raw"
))
165
op.
set
(
"raw"
, 1);
166
}
main
spectgen_main.cc
Generated on Wed Dec 24 2014 09:16:35 for Edinburgh Speech Tools by
1.8.3.1