1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
|
#!/usr/bin/python2
### PicoSpeaker
### Written by Kyle
### A small program that speaks text on its command line or standard input using SVox Pico and Sox
### Speech rate, pitch, volume and language can be specified, and output can be saved to many file formats
### This program is free and unencumbered software released into the public domain. See UNLICENSE for details.
### SVox Pico and Sox are covered by their own open source licenses and copyrights,
### and are therefore not covered by the UNLICENSE included with PicoSpeaker.
from sys import argv, stdin, stderr, exit
from subprocess import call, Popen
from os import environ as env, remove
from time import sleep
# help and version tuples
version = (
'PicoSpeaker 0.6.2',
'Written by Kyle',
'This program is free and unencumbered software released into the public domain.',
'See the included UNLICENSE file for details.')
help = (
'Usage:',
argv[0] + ' [options] <text>',
'',
'Options:',
'-l|--language <language> Language to speak: (default is en-US)',
'-v|--volume <number> Output volume: (default is 1.0)',
'-r|--rate <number> Rate of speech from -90 to 9900: (default is 0)',
'-p|--pitch <number> Voice pitch from -79 to 39: (default is 0)',
'-o|--output <file> Output to the specified file: (default is sound card output)',
'-c|--compress|-q|--quality <number> Compression/quality level of output file, depends on file type',
' This option causes an error if no output file is specified.',
'-t|--type <type> Save output file as <type>. Only needed if saving with a nonstandard extension.',
' This option causes an error if no output file is specified.',
'-V|--version Print version information',
'-h|--help|-u|--usage Print this help message')
# Spoken text is stored in this temp file; unfortunately, pico2wave doesn't pipe
temp = '/tmp/picospeaker-' + env['USER'] + '.wav'
def parse ():
'Parse the command line and return settings and text to be spoken in a dictionary object'
settings = {} #Voice parameters and command line options, the object returned by the function
# The parser loop, which populates the settings object and handles usage and version options
skip = False #set True to skip the next item in the sequence.
for opt in range(1,len(argv)):
if skip:
skip = False
continue
if ( argv[opt] == '-V' ) or ( argv[opt] == '--version' ):
for line in version: stderr.write(line + '\n')
exit(0)
elif ( argv[opt] == '-h' ) or ( argv[opt] == '--help' ) \
or ( argv[opt] == '-u' ) or ( argv[opt] == '--usage' ):
for line in version: stderr.write(line + '\n')
stderr.write('\n')
for line in help: stderr.write(line + '\n')
exit(0)
elif ( argv[opt] == '-l' ) or ( argv[opt] == '--language' ):
languages = ('en-US', 'en-GB', 'de-DE', 'es-ES', 'fr-FR', 'it-IT')
if ( argv[opt+1] in languages ):
settings['language'] = argv[opt+1]
else:
stderr.write('Language ' + argv[opt+1] + ' is currently not available.\n')
stderr.write('Available languages are ' + ', '.join(languages[:-1]) + ' and ' + languages[-1] + '.\n')
exit(1)
skip = True
continue
elif ( argv[opt] == '-p' ) or ( argv[opt] == '--pitch' ):
try:
if ( int(argv[opt+1]) in range(-79, 40) ):
settings['pitch'] = argv[opt+1]
else: raise ValueError
except ValueError:
stderr.write('Pitch must be a whole number from -79 to 39.\n')
exit(1)
skip = True
continue
elif ( argv[opt] == '-r' ) or ( argv[opt] == '--rate' ):
try:
if ( int(argv[opt+1]) in range(-90, 9901) ):
settings['rate'] = argv[opt+1]
else: raise ValueError
except ValueError:
stderr.write('Rate must be a whole number from -90 to 9900.\n')
exit(1)
skip = True
continue
elif ( argv[opt] == '-v' ) or ( argv[opt] == '--volume' ):
try:
float(argv[opt+1])
except ValueError:
stderr.write('Volume must be a number.\n')
exit(1)
settings['volume'] = argv[opt+1]
skip = True
continue
elif ( argv[opt] == '-o' ) or ( argv[opt] == '--output' ):
# TODO: Check for settings['filetype'] if output file extension would be unrecognized by Sox: may not be possible here
settings['output'] = argv[opt+1]
skip = True
continue
elif ( argv[opt] == '-c' ) or ( argv[opt] == '--compress' ) \
or ( argv[opt] == '-q' ) or ( argv[opt] == '--quality' ):
try:
float(argv[opt+1]) #must be a number
except ValueError:
stderr.write('Compression/quality level must be a number.\n')
exit(1)
settings['compression'] = argv[opt+1]
skip = True
continue
elif ( argv[opt] == '-t' ) or ( argv[opt] == '--type' ):
settings['filetype'] = argv[opt+1]
skip = True
continue
else:
# First, die with an error if compression and/or type are set but no output file is specified
if ( ( 'compression' in settings ) or ( 'filetype' in settings ) ) and ( 'output' not in settings ):
stderr.write('You must specify the output file.\n')
exit(1)
# Now the text can be added to the settings object and the loop can be broken
settings['text'] = ' '.join(argv[opt:])
break
return settings
def tts():
'convert text to speech data and store it in a temporary file using the pico2wave utility from SVox Pico'
command = ['pico2wave', '-w', temp]
if ( 'language' in settings ): command += ['-l', settings['language']]
command += ['--', settings['text']]
try:
call(command)
except OSError:
stderr.write('FIXME: text is too large.\n')
exit(1)
def speaker():
'speaks the text, or saves it if an output file was specified on the command line'
command = ['/usr/bin/play', '-q']
if ( 'volume' in settings ): command += ['-v', settings['volume']]
command.append(temp)
if ( 'output' in settings ):
command[0] = '/usr/bin/sox'
del command[1]
if ( 'filetype' in settings ): command += ['-t', settings['filetype']]
if ( 'compression' in settings ): command += ['-C', settings['compression']]
command.append(settings['output'])
if ( 'pitch' in settings ): command += ['gain', '-0.15', 'pitch', str(float(settings['pitch'])*100)]
if ( 'rate' in settings ): command += ['gain', '-0.1', 'tempo', '-s', str(1+float(settings['rate'])/100)]
speak = Popen(command)
sleep(0.1) # the temp file should be open by now
# The temp file can be removed as soon as it is opened in case PicoSpeaker is killed while speaking
remove(temp)
speak.wait() # Don't leave this function until the command is completed
try:
settings = parse()
if ( 'text' not in settings ):
settings['text'] = stdin.read()
tts()
speaker()
except KeyboardInterrupt:
stderr.write('Keyboard interrupt received. Cleaning up.\n')
try: remove(temp) # The temp file may not have been removed yet
except OSError: pass # The file doesn't exist and therefore doesn't need to be removed
|