def _synthesize_single_helper(self, text, voice_code, output_file_path=None):
"""
This is an helper function to synthesize a single text fragment via Python call.
The caller can choose whether the output file should be written to disk or not.
:rtype: tuple (result, (duration, sample_rate, encoding, data))
"""
#
# NOTE in this example, we assume that the Speect voice data files
# are located in the same directory of this .py source file
# and that the voice JSON file is called "voice.json"
#
# NOTE the voice_code value is ignored in this example,
# but in general one might select a voice file to load,
# depending on voice_code
#
voice_json_path = gf.safe_str(gf.absolute_path("voice.json", __file__))
voice = speect.SVoice(voice_json_path)
utt = voice.synth(text)
audio = utt.features["audio"]
if output_file_path is not None:
audio.save_riff(gf.safe_str(output_file_path))
# get length and data using speect Python API
waveform = audio.get_audio_waveform()
audio_sample_rate = int(waveform["samplerate"])
audio_length = TimeValue(audio.num_samples() / audio_sample_rate)
audio_format = "pcm16"
audio_samples = numpy.fromstring(waveform["samples"], dtype=numpy.int16).astype("float64") / 32768
# return data
return (audio_length, audio_sample_rate, audio_format, audio_samples)
def test_safe_str(self):
tests = [
(u"", ""),
(u"foo", "foo"),
(u"foà", "foà"),
]
self.assertIsNone(gf.safe_str(None))
for test in tests:
self.assertEqual(gf.safe_str(test[0]), test[1])
def test_safe_str(self):
tests = [
(u"", ""),
(u"foo", "foo"),
(u"foà", "foà"),
]
self.assertIsNone(gf.safe_str(None))
for test in tests:
self.assertEqual(gf.safe_str(test[0]), test[1])
def _synthesize_single_helper(self,
text,
voice_code,
output_file_path=None):
"""
This is an helper function to synthesize a single text fragment via Python call.
The caller can choose whether the output file should be written to disk or not.
:rtype: tuple (result, (duration, sample_rate, encoding, data))
"""
#
# NOTE in this example, we assume that the Speect voice data files
# are located in the same directory of this .py source file
# and that the voice JSON file is called "voice.json"
#
# NOTE the voice_code value is ignored in this example,
# but in general one might select a voice file to load,
# depending on voice_code
#
voice_json_path = gf.safe_str(gf.absolute_path("voice.json", __file__))
voice = speect.SVoice(voice_json_path)
utt = voice.synth(text)
audio = utt.features["audio"]
if output_file_path is not None:
audio.save_riff(gf.safe_str(output_file_path))
# get length and data using speect Python API
waveform = audio.get_audio_waveform()
audio_sample_rate = int(waveform["samplerate"])
audio_length = TimeValue(audio.num_samples() / audio_sample_rate)
audio_format = "pcm16"
audio_samples = numpy.fromstring(
waveform["samples"], dtype=numpy.int16).astype("float64") / 32768
# return data
return (audio_length, audio_sample_rate, audio_format, audio_samples)
def __str__(self):
return gf.safe_str(self.__unicode__())
def __str__(self):
return gf.safe_str(self.__unicode__())
def _synthesize_single_python_helper(
self,
text,
voice_code,
output_file_path=None,
return_audio_data=True
):
"""
This is an helper function to synthesize a single text fragment via a Python call.
If ``output_file_path`` is ``None``,
the audio data will not persist to file at the end of the method.
:rtype: tuple (result, (duration, sample_rate, encoding, data))
"""
# return zero if text is the empty string
if len(text) == 0:
#
# NOTE values of sample_rate, encoding, data
# do not matter if the duration is 0.000,
# so set them to None instead of the more precise:
# return (True, (TimeValue("0.000"), 16000, "pcm_s16le", numpy.array([])))
#
self.log(u"len(text) is zero: returning 0.000")
return (True, (TimeValue("0.000"), None, None, None))
#
# NOTE in this example, we assume that the Speect voice data files
# are located in the same directory of this .py source file
# and that the voice JSON file is called "voice.json"
#
# NOTE the voice_code value is ignored in this example,
# since we have only one TTS voice,
# but in general one might select a voice file to load,
# depending on voice_code;
# in fact, we could have created the ``voice`` object
# only once, in the constructor, instead of creating it
# each time this function is invoked,
# achieving slightly faster synthesis
#
voice_json_path = gf.safe_str(gf.absolute_path("voice.json", __file__))
voice = speect.SVoice(voice_json_path)
utt = voice.synth(text)
audio = utt.features["audio"]
if output_file_path is None:
self.log(u"output_file_path is None => not saving to file")
else:
self.log(u"output_file_path is not None => saving to file...")
# NOTE apparently, save_riff needs the path to be a byte string
audio.save_riff(gf.safe_str(output_file_path))
self.log(u"output_file_path is not None => saving to file... done")
# return immediately if returning audio data is not needed
if not return_audio_data:
self.log(u"return_audio_data is True => return immediately")
return (True, None)
# get length and data using speect Python API
self.log(u"return_audio_data is True => read and return audio data")
waveform = audio.get_audio_waveform()
audio_sample_rate = int(waveform["samplerate"])
audio_length = TimeValue(audio.num_samples() / audio_sample_rate)
audio_format = "pcm16"
audio_samples = numpy.fromstring(
waveform["samples"],
dtype=numpy.int16
).astype("float64") / 32768
return (True, (
audio_length,
audio_sample_rate,
audio_format,
audio_samples
))
