transcribed = [
x[0].upper()
for x in sorted(results, key=lambda x: x[1], reverse=True)
]
self._logger.info('Transcribed: %r', transcribed)
return transcribed
def _get_response(self, data):
headers = {
'authorization': 'Bearer %s' % self.token,
'accept': 'application/json',
'content-type': 'audio/wav'
}
return requests.post('https://api.att.com/speech/v3/speechToText',
data=data,
headers=headers)
@classmethod
def is_available(cls):
return diagnose.check_network_connection()
def is_valid():
'''
Module validator.
'''
return True
initiator = _module_getter(AttSTT)
]
cmd = [str(x) for x in cmd]
self._logger.debug('Executing: %r', cmd)
with tempfile.SpooledTemporaryFile() as out_f:
with tempfile.SpooledTemporaryFile() as err_f:
subprocess.call(cmd, stdin=fp, stdout=out_f, stderr=err_f)
out_f.seek(0)
results = [(int(i), text)
for i, text in self._pattern.findall(out_f.read())]
transcribed = [
text for i, text in sorted(results, key=lambda x: x[0]) if text
]
if not transcribed:
transcribed.append('')
self._logger.info('Transcribed: %r', transcribed)
return transcribed
@classmethod
def is_available(cls):
return diagnose.check_executable('julius')
def is_valid():
'''
Module validator.
'''
return True
initiator = _module_getter(JuliusSTT)
def say(self, phrase, *args):
self._logger.debug("Saying '%s' with '%s'", phrase, self.SLUG)
with tempfile.NamedTemporaryFile(suffix='.wav', delete=False) as f:
fname = f.name
cmd = [
'espeak', '-v', self.voice, '-p', self.pitch_adjustment, '-s',
self.words_per_minute, '-w', fname, phrase
]
cmd = [str(x) for x in cmd]
self._logger.debug('Executing %s',
' '.join([pipes.quote(arg) for arg in cmd]))
with tempfile.TemporaryFile() as f:
subprocess.call(cmd, stdout=f, stderr=f)
f.seek(0)
output = f.read()
if output:
self._logger.debug("Output was: '%s'", output)
self.play(fname)
os.remove(fname)
def is_valid():
'''
Validator.
'''
return True
initiator = _module_getter(EspeakTTS)
self._logger.warning('Status 403 is probably caused by invalid credentials.')
return []
r.encoding = 'utf-8'
results = []
try:
response = r.json()
if not response['results']:
raise ValueError('Nothing has been transcribed.')
results = tuple([alt['transcript'].strip().upper() for alt in response['results'][0]['alternatives']])
self._logger.info('Transcribed: %r', results)
except ValueError as e:
self._logger.critical('Empty response: %s', e.args[0])
except (KeyError, IndexError):
self._logger.critical('Cannot parse response.', exc_info=True)
return results
@classmethod
def is_available(cls):
return diagnose.check_network_connection()
def is_valid():
'''
Module validator.
'''
return True
initiator = _module_getter(WatsonSTT)
# FIXME: Can't use the Decoder.decode_raw() here, because
# pocketsphinx segfaults with tempfile.SpooledTemporaryFile()
data = fp.read()
self._decoder.start_utt()
self._decoder.process_raw(data, False, True)
self._decoder.end_utt()
result = self._decoder.get_hyp()
with open(self._logfile, 'r+') as f:
for line in f:
self._logger.debug(line.strip())
f.truncate()
transcribed = [result[0]]
self._logger.info('Transcribed: %r', transcribed)
return transcribed
@classmethod
def is_available(cls):
return diagnose.check_python_import('pocketsphinx')
def is_valid():
'''
Module validator.
'''
return True
initiator = _module_getter(PocketSphinxSTT)
if 'speech_rate' in profile['ivona-tts']:
config['speech_rate'] = profile['ivona-tts']['speech_rate']
if 'sentence_break' in profile['ivona-tts']:
config['sentence_break'] = profile['ivona-tts']['sentence_break']
return config
@classmethod
def is_available(cls):
return (super(cls, cls).is_available() and
diagnose.check_python_import('pyvona') and
diagnose.check_network_connection())
def say(self, phrase, *args):
self._logger.debug("Saying '%s' with '%s'", phrase, self.SLUG)
with tempfile.NamedTemporaryFile(suffix='.mp3', delete=False) as f:
tmpfile = f.name
self._pyvonavoice.fetch_voice(phrase, tmpfile)
self.play_mp3(tmpfile)
os.remove(tmpfile)
def is_valid():
'''
Validator.
'''
return HAS_PYVONA
initiator = _module_getter(IvonaTTS)
cmd = ['say', str(phrase)]
self._logger.debug('Executing %s',
' '.join([pipes.quote(arg) for arg in cmd]))
with tempfile.TemporaryFile() as f:
subprocess.call(cmd, stdout=f, stderr=f)
f.seek(0)
output = f.read()
if output:
self._logger.debug("Output was: '%s'", output)
def play(self, filename):
cmd = ['afplay', str(filename)]
self._logger.debug('Executing %s',
' '.join([pipes.quote(arg) for arg in cmd]))
with tempfile.TemporaryFile() as f:
subprocess.call(cmd, stdout=f, stderr=f)
f.seek(0)
output = f.read()
if output:
self._logger.debug("Output was: '%s'", output)
def is_valid():
'''
Validator.
'''
return True
initiator = _module_getter(MacOSXTTS)
def languages(self):
langs = [
'af', 'sq', 'ar', 'hy', 'ca', 'zh-CN', 'zh-TW', 'hr', 'cs', 'da',
'nl', 'en', 'eo', 'fi', 'fr', 'de', 'el', 'ht', 'hi', 'hu', 'is',
'id', 'it', 'ja', 'ko', 'la', 'lv', 'mk', 'no', 'pl', 'pt', 'ro',
'ru', 'sr', 'sk', 'es', 'sw', 'sv', 'ta', 'th', 'tr', 'vi', 'cy'
]
return langs
def say(self, phrase, *args):
self._logger.debug("Saying '%s' with '%s'", phrase, self.SLUG)
if self.language not in self.languages:
raise ValueError("Language '%s' not supported by '%s'",
self.language, self.SLUG)
tts = gtts.gTTS(text=phrase, lang=self.language)
with tempfile.NamedTemporaryFile(suffix='.mp3', delete=False) as f:
tmpfile = f.name
tts.save(tmpfile)
self.play_mp3(tmpfile)
os.remove(tmpfile)
def is_valid():
'''
Validator.
'''
return HAS_GTTS
initiator = _module_getter(GoogleTTS)
(self.language, self.SLUG))
if self.voice not in self.voices:
raise ValueError("Voice '%s' not supported by '%s'" %
(self.voice, self.SLUG))
query = {
'OUTPUT_TYPE': 'AUDIO',
'AUDIO': 'WAVE_FILE',
'INPUT_TYPE': 'TEXT',
'INPUT_TEXT': phrase,
'LOCALE': self.language,
'VOICE': self.voice
}
r = self.session.get(self._makeurl('/process', query=query))
with tempfile.NamedTemporaryFile(suffix='.wav', delete=False) as f:
f.write(r.content)
tmpfile = f.name
self.play(tmpfile)
os.remove(tmpfile)
def is_valid():
'''
Validator.
'''
return True
initiator = _module_getter(MaryTTS)
def say(self, phrase, *args):
self._logger.debug("Saying '%s' with '%s'", phrase, self.SLUG)
cmd = ['flite']
if self.voice:
cmd.extend(['-voice', self.voice])
cmd.extend(['-t', phrase])
with tempfile.NamedTemporaryFile(suffix='.wav', delete=False) as f:
fname = f.name
cmd.append(fname)
with tempfile.SpooledTemporaryFile() as out_f:
self._logger.debug('Executing %s',
' '.join([pipes.quote(arg) for arg in cmd]))
subprocess.call(cmd, stdout=out_f, stderr=out_f)
out_f.seek(0)
output = out_f.read().strip()
if output:
self._logger.debug("Output was: '%s'", output)
self.play(fname)
os.remove(fname)
def is_valid():
'''
Validator.
'''
return True
initiator = _module_getter(FliteTTS)
if len(response['result']) == 0:
# Response result is empty
raise ValueError('Nothing has been transcribed.')
results = [alt['transcript'] for alt
in response['result'][0]['alternative']]
except ValueError as e:
self._logger.warning('Empty response: %s', e.args[0])
results = []
except (KeyError, IndexError):
self._logger.warning('Cannot parse response.', exc_info=True)
results = []
else:
# Convert all results to uppercase
results = tuple(result.upper() for result in results)
self._logger.info('Transcribed: %r', results)
return results
@classmethod
def is_available(cls):
return diagnose.check_network_connection()
def is_valid():
'''
Module validator.
'''
return True
initiator = _module_getter(GoogleSTT)
text = r.json()['_text']
except requests.exceptions.HTTPError:
self._logger.critical('Request failed with response: %r',
r.text,
exc_info=True)
return []
except requests.exceptions.RequestException:
self._logger.critical('Request failed.', exc_info=True)
return []
except ValueError as e:
self._logger.critical('Cannot parse response: %s',
e.args[0])
return []
except KeyError:
self._logger.critical('Cannot parse response.',
exc_info=True)
return []
else:
transcribed = []
if text:
transcribed.append(text.upper())
self._logger.info('Transcribed: %r', transcribed)
return transcribed
@classmethod
def is_available(cls):
return diagnose.check_network_connection()
initiator = _module_getter(WitAiSTT)
class DummyTTS(AbstractTTSEngine):
"""
Dummy TTS engine that logs phrases with INFO level instead of synthesizing
speech.
"""
SLUG = "dummy-tts"
@classmethod
def is_available(cls):
return True
def say(self, phrase, *args):
self._logger.info(phrase)
def play(self, filename):
self._logger.debug("Playback of file '%s' requested")
pass
def is_valid():
'''
Validator.
'''
return True
initiator = _module_getter(DummyTTS)
return False
def say(self, phrase, *args):
self._logger.debug("Saying '%s' with '%s'", phrase, self.SLUG)
cmd = ['text2wave']
with tempfile.NamedTemporaryFile(suffix='.wav') as out_f:
with tempfile.SpooledTemporaryFile() as in_f:
in_f.write(phrase)
in_f.seek(0)
with tempfile.SpooledTemporaryFile() as err_f:
self._logger.debug('Executing %s',
' '.join([pipes.quote(arg)
for arg in cmd]))
subprocess.call(cmd, stdin=in_f, stdout=out_f,
stderr=err_f)
err_f.seek(0)
output = err_f.read()
if output:
self._logger.debug("Output was: '%s'", output)
self.play(out_f.name)
def is_valid():
'''
Validator.
'''
return True
initiator = _module_getter(FestivalTTS)
def say(self, phrase, *args):
self._logger.debug("Saying '%s' with '%s'", phrase, self.SLUG)
with tempfile.NamedTemporaryFile(suffix='.wav', delete=False) as f:
fname = f.name
cmd = ['pico2wave', '--wave', fname]
if self.language not in self.languages:
raise ValueError("Language '%s' not supported by '%s'",
self.language, self.SLUG)
cmd.extend(['-l', self.language])
cmd.append(phrase)
self._logger.debug('Executing %s',
' '.join([pipes.quote(arg) for arg in cmd]))
with tempfile.TemporaryFile() as f:
subprocess.call(cmd, stdout=f, stderr=f)
f.seek(0)
output = f.read()
if output:
self._logger.debug("Output was: '%s'", output)
self.play(fname)
os.remove(fname)
def is_valid():
'''
Validator.
'''
return True
initiator = _module_getter(PicoTTS)