Edinburgh Speech Tools
2.4-release
Loading...
Searching...
No Matches
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
97
EST_Wave
sig
;
98
EST_Track
spec
;
99
100
parse_command_line
101
(
argc
,
argv
,
102
EST_String
(
"[input file] -o [output file]\n"
)+
103
"Summary: make spectrogram\n"
+
104
"use \"-\" to make input and output files stdin/out\n"
+
105
"-h Options help\n"
+
106
options_wave_input()+
107
"\n"
+
108
options_track_output()+
109
"-shift <float> frame spacing in seconds for fixed frame analysis. This \n"
110
" doesn't have to be the same as the output file spacing - the \n"
111
" S option can be used to resample the track before saving \n"
112
" default: "
+ftoString(DEFAULT_FRAME_SIZE) +
"\n\n"
113
"-length <float> input frame length in milliseconds\n"
+
114
"-sr <float> range in which output values should lie\n"
+
115
"-slow slow FFT code\n"
+
116
"-w <float> white cut off (0.0 to 1.0)\n"
+
117
"-b <float> black cut off (0.0 to 1.0)\n"
+
118
"-raw Don't perform any scaling\n"
+
119
"-order <int> cepstral order\n"
,
files
,
al
);
120
121
out_file =
al
.present(
"-o"
) ?
al
.val(
"-o"
) : (
EST_String
)
"-"
;
122
set_options(op,
al
);
123
124
if
(read_wave(
sig
,
files
.first(),
al
) != format_ok)
125
exit
(-1);
126
127
make_spectrogram(
sig
,
spec
, op);
128
129
spec
.save(out_file,
al
.val(
"-otype"
, 0));
130
131
return
0;
132
}
133
134
void
set_options(
EST_Features
&op,
EST_Option
&
al
)
135
{
136
op.
set
(
"frame_shift"
, DEFAULT_FRAME_SIZE);
137
op.
set
(
"frame_length"
, DEFAULT_FRAME_LENGTH);
138
op.
set
(
"preemph"
, DEFAULT_PREEMPH);
139
op.
set
(
"frame_order"
, DEFAULT_ORDER);
140
141
if
(
al
.present(
"-shift"
))
142
op.
set
(
"frame_shift"
,
al
.fval(
"-shift"
));
143
144
if
(
al
.present(
"-length"
))
145
op.
set
(
"frame_length"
,
al
.fval(
"-length"
));
146
147
if
(
al
.present(
"-order"
))
148
op.
set
(
"frame_order"
,
al
.fval(
"-order"
));
149
150
if
(
al
.present(
"-sr"
))
151
op.
set
(
"sp_range"
,
al
.fval(
"-sr"
));
152
153
if
(
al
.present(
"-w"
))
154
op.
set
(
"sp_wcut"
,
al
.fval(
"-w"
));
155
156
if
(
al
.present(
"-b"
))
157
op.
set
(
"sp_bcut"
,
al
.fval(
"-b"
));
158
159
if
(
al
.present(
"-preemph"
))
160
op.
set
(
"preemph"
,
al
.fval(
"-preemph"
, 1));
161
162
if
(
al
.present(
"-raw"
))
163
op.
set
(
"raw"
, 1);
164
}
EST_Features
Definition
EST_Features.h:62
EST_Features::set
void set(const EST_String &name, int ival)
Definition
EST_Features.h:185
EST_Hash_Pair
Definition
EST_THash.h:75
EST_Option
Definition
EST_Option.h:50
EST_String
Definition
EST_String.h:70
EST_TList
Definition
EST_TList.h:109
EST_Track
Definition
EST_Track.h:89
EST_Wave
Definition
EST_Wave.h:65
main
spectgen_main.cc
Generated on Thu Mar 28 2024 14:18:19 for Edinburgh Speech Tools by
1.9.8