def _synthesize_multiple_python(self, text_file, output_file_path, quit_after=None, backwards=False):
"""
Synthesize multiple text fragments, via Python call.
Return a tuple (anchors, total_time, num_chars).
:rtype: (bool, (list, TimeValue, int))
"""
#
# generating wave data for each fragment,
# and concatenating them together
#
self.log(u"Calling TTS engine via Python...")
try:
# open output file
output_file = AudioFile(rconf=self.rconf, logger=self.logger)
output_file.audio_format = "pcm16"
output_file.audio_channels = 1
output_file.audio_sample_rate = self.SAMPLE_RATE
# create output
anchors = []
current_time = TimeValue("0.000")
num = 0
num_chars = 0
fragments = text_file.fragments
if backwards:
fragments = fragments[::-1]
for fragment in fragments:
# language to voice code
voice_code = self._language_to_voice_code(fragment.language)
# synthesize and get the duration of the output file
self.log([u"Synthesizing fragment %d", num])
duration, sr_nu, enc_nu, data = self._synthesize_single_helper(
text=(fragment.filtered_text + u" "),
voice_code=voice_code
)
# store for later output
anchors.append([current_time, fragment.identifier, fragment.text])
# increase the character counter
num_chars += fragment.characters
# append new data
self.log([u"Fragment %d starts at: %.3f", num, current_time])
if duration > 0:
self.log([u"Fragment %d duration: %.3f", num, duration])
current_time += duration
# if backwards, we append the data reversed
output_file.add_samples(data, reverse=backwards)
else:
self.log([u"Fragment %d has zero duration", num])
# increment fragment counter
num += 1
# check if we must stop synthesizing because we have enough audio
if (quit_after is not None) and (current_time > quit_after):
self.log([u"Quitting after reached duration %.3f", current_time])
break
# if backwards, we need to reverse the audio samples again
if backwards:
output_file.reverse()
# write output file
self.log([u"Writing audio file '%s'", output_file_path])
output_file.write(file_path=output_file_path)
except Exception as exc:
self.log_exc(u"Unexpected exception while calling TTS engine via Python", exc, None, type(exc))
return (False, None)
# return output
# NOTE anchors do not make sense if backwards
self.log([u"Returning %d time anchors", len(anchors)])
self.log([u"Current time %.3f", current_time])
self.log([u"Synthesized %d characters", num_chars])
self.log(u"Calling TTS engine via Python... done")
return (True, (anchors, current_time, num_chars))
def _cut_head_tail(self, audio_file_path):
"""
Set the audio file head or tail,
suitably cutting the audio file on disk,
and setting the corresponding parameters in the task configuration.
Return a success bool flag
"""
self._log("Setting head and/or tail")
try:
configuration = self.task.configuration
head_length = configuration.is_audio_file_head_length
process_length = configuration.is_audio_file_process_length
detect_head_min = configuration.is_audio_file_detect_head_min
detect_head_max = configuration.is_audio_file_detect_head_max
detect_tail_min = configuration.is_audio_file_detect_tail_min
detect_tail_max = configuration.is_audio_file_detect_tail_max
# explicit head or process?
explicit = (head_length is not None) or (process_length
is not None)
# at least one detect parameter?
detect = ((detect_head_min is not None)
or (detect_head_max is not None)
or (detect_tail_min is not None)
or (detect_tail_max is not None))
if explicit or detect:
# we need to load the audio data
audio_file = AudioFile(audio_file_path, logger=self.logger)
audio_file.load_data()
if explicit:
self._log("Explicit head or process")
else:
self._log(
"No explicit head or process => detecting head/tail")
head = 0.0
if (detect_head_min is not None) or (detect_head_max
is not None):
self._log("Detecting head...")
detect_head_min = gf.safe_float(
detect_head_min, gc.SD_MIN_HEAD_LENGTH)
detect_head_max = gf.safe_float(
detect_head_max, gc.SD_MAX_HEAD_LENGTH)
self._log(["detect_head_min is %.3f", detect_head_min])
self._log(["detect_head_max is %.3f", detect_head_max])
sd = SD(audio_file,
self.task.text_file,
logger=self.logger)
head = sd.detect_head(detect_head_min, detect_head_max)
self._log(["Detected head: %.3f", head])
tail = 0.0
if (detect_tail_min is not None) or (detect_tail_max
is not None):
self._log("Detecting tail...")
detect_tail_max = gf.safe_float(
detect_tail_max, gc.SD_MAX_TAIL_LENGTH)
detect_tail_min = gf.safe_float(
detect_tail_min, gc.SD_MIN_TAIL_LENGTH)
self._log(["detect_tail_min is %.3f", detect_tail_min])
self._log(["detect_tail_max is %.3f", detect_tail_max])
sd = SD(audio_file,
self.task.text_file,
logger=self.logger)
tail = sd.detect_tail(detect_tail_min, detect_tail_max)
self._log(["Detected tail: %.3f", tail])
# sanity check
head_length = max(0, head)
process_length = max(0,
audio_file.audio_length - tail - head)
# we need to set these values
# in the config object for later use
self.task.configuration.is_audio_file_head_length = head_length
self.task.configuration.is_audio_file_process_length = process_length
self._log(["Set head_length: %.3f", head_length])
self._log(["Set process_length: %.3f", process_length])
if head_length is not None:
# in case we are reading from config object
head_length = float(head_length)
if process_length is not None:
# in case we are reading from config object
process_length = float(process_length)
# note that str() is necessary, as one might be None
self._log(
["is_audio_file_head_length is %s",
str(head_length)])
self._log([
"is_audio_file_process_length is %s",
str(process_length)
])
self._log("Trimming audio data...")
audio_file.trim(head_length, process_length)
self._log("Trimming audio data... done")
self._log("Writing audio file...")
audio_file.write(audio_file_path)
self._log("Writing audio file... done")
audio_file.clear_data()
else:
# nothing to do
self._log("No explicit head/process or detect head/tail")
self._log("Setting head and/or tail: succeeded")
return True
except Exception as e:
self._log("Setting head and/or tail: failed")
self._log(["Message: %s", str(e)])
return False
def _synthesize_multiple_python(self,
text_file,
output_file_path,
quit_after=None,
backwards=False):
"""
Synthesize multiple text fragments, via Python call.
Return a tuple (anchors, total_time, num_chars).
:rtype: (bool, (list, TimeValue, int))
"""
#
# generating wave data for each fragment,
# and concatenating them together
#
self.log(u"Calling TTS engine via Python...")
try:
# open output file
output_file = AudioFile(rconf=self.rconf, logger=self.logger)
output_file.audio_format = "pcm16"
output_file.audio_channels = 1
output_file.audio_sample_rate = self.SAMPLE_RATE
# create output
anchors = []
current_time = TimeValue("0.000")
num = 0
num_chars = 0
fragments = text_file.fragments
if backwards:
fragments = fragments[::-1]
for fragment in fragments:
# language to voice code
voice_code = self._language_to_voice_code(fragment.language)
# synthesize and get the duration of the output file
self.log([u"Synthesizing fragment %d", num])
duration, sr_nu, enc_nu, data = self._synthesize_single_helper(
text=(fragment.filtered_text + u" "),
voice_code=voice_code)
# store for later output
anchors.append(
[current_time, fragment.identifier, fragment.text])
# increase the character counter
num_chars += fragment.characters
# append new data
self.log([u"Fragment %d starts at: %.3f", num, current_time])
if duration > 0:
self.log([u"Fragment %d duration: %.3f", num, duration])
current_time += duration
# if backwards, we append the data reversed
output_file.add_samples(data, reverse=backwards)
else:
self.log([u"Fragment %d has zero duration", num])
# increment fragment counter
num += 1
# check if we must stop synthesizing because we have enough audio
if (quit_after is not None) and (current_time > quit_after):
self.log([
u"Quitting after reached duration %.3f", current_time
])
break
# if backwards, we need to reverse the audio samples again
if backwards:
output_file.reverse()
# write output file
self.log([u"Writing audio file '%s'", output_file_path])
output_file.write(file_path=output_file_path)
except Exception as exc:
self.log_exc(
u"Unexpected exception while calling TTS engine via Python",
exc, None, type(exc))
return (False, None)
# return output
# NOTE anchors do not make sense if backwards
self.log([u"Returning %d time anchors", len(anchors)])
self.log([u"Current time %.3f", current_time])
self.log([u"Synthesized %d characters", num_chars])
self.log(u"Calling TTS engine via Python... done")
return (True, (anchors, current_time, num_chars))
def _synthesize_multiple_generic(self,
helper_function,
text_file,
output_file_path,
quit_after=None,
backwards=False):
"""
Synthesize multiple fragments, generic function.
The ``helper_function`` is a function that takes parameters
``(text, voice_code, output_file_path)``
and returns a tuple
``(result, (audio_length, audio_sample_rate, audio_format, audio_samples))``.
:rtype: tuple (result, (anchors, current_time, num_chars))
"""
self.log(u"Calling TTS engine using multiple generic function...")
# get sample rate and codec
self.log(u"Determining codec and sample rate...")
if (self.OUTPUT_AUDIO_FORMAT is None) or (len(self.OUTPUT_AUDIO_FORMAT)
!= 3):
self.log(u"Determining codec and sample rate with dummy text...")
succeeded, data = helper_function(
text=u"Dummy text to get sample_rate",
voice_code=self._language_to_voice_code(self.DEFAULT_LANGUAGE),
output_file_path=None)
if not succeeded:
self.log_crit(
u"An unexpected error occurred in helper_function")
return (False, None)
du_nu, sample_rate, codec, da_nu = data
self.log(
u"Determining codec and sample rate with dummy text... done")
else:
self.log(u"Reading codec and sample rate from OUTPUT_AUDIO_FORMAT")
codec, channels_nu, sample_rate = self.OUTPUT_AUDIO_FORMAT
self.log(u"Determining codec and sample rate... done")
self.log([u" codec: %s", codec])
self.log([u" sample rate: %d", sample_rate])
# open output file
output_file = AudioFile(rconf=self.rconf, logger=self.logger)
output_file.audio_format = codec
output_file.audio_channels = 1
output_file.audio_sample_rate = sample_rate
# create output
anchors = []
current_time = TimeValue("0.000")
num_chars = 0
fragments = text_file.fragments
if backwards:
fragments = fragments[::-1]
loop_function = self._loop_use_cache if self.use_cache else self._loop_no_cache
for num, fragment in enumerate(fragments):
succeeded, data = loop_function(helper_function=helper_function,
num=num,
fragment=fragment)
if not succeeded:
self.log_crit(u"An unexpected error occurred in loop_function")
return (False, None)
duration, sr_nu, enc_nu, samples = data
# store for later output
anchors.append([current_time, fragment.identifier, fragment.text])
# increase the character counter
num_chars += fragment.characters
# concatenate new samples
self.log([u"Fragment %d starts at: %.3f", num, current_time])
if duration > 0:
self.log([u"Fragment %d duration: %.3f", num, duration])
current_time += duration
output_file.add_samples(samples, reverse=backwards)
else:
self.log([u"Fragment %d has zero duration", num])
# check if we must stop synthesizing because we have enough audio
if (quit_after is not None) and (current_time > quit_after):
self.log(
[u"Quitting after reached duration %.3f", current_time])
break
# minimize memory
self.log(u"Minimizing memory...")
output_file.minimize_memory()
self.log(u"Minimizing memory... done")
# if backwards, we need to reverse the audio samples again
if backwards:
self.log(u"Reversing audio samples...")
output_file.reverse()
self.log(u"Reversing audio samples... done")
# write output file
self.log([u"Writing audio file '%s'", output_file_path])
output_file.write(file_path=output_file_path)
# return output
if backwards:
self.log_warn(
u"Please note that anchor time values do not make sense since backwards=True"
)
self.log([u"Returning %d time anchors", len(anchors)])
self.log([u"Current time %.3f", current_time])
self.log([u"Synthesized %d characters", num_chars])
self.log(u"Calling TTS engine using multiple generic function... done")
return (True, (anchors, current_time, num_chars))
def _synthesize_multiple_subprocess(self, text_file, output_file_path, quit_after=None, backwards=False):
"""
Synthesize multiple fragments via ``subprocess``.
:rtype: tuple (result, (anchors, current_time, num_chars))
"""
def synthesize_and_clean(text, voice_code):
"""
Synthesize a single fragment via subprocess,
and immediately remove the temporary file.
:rtype: tuple (duration, sample_rate, encoding, samples)
"""
self.log(u"Synthesizing text...")
handler, tmp_destination = gf.tmp_file(suffix=u".wav", root=self.rconf[RuntimeConfiguration.TMP_PATH])
result, data = self._synthesize_single_subprocess(
text=(text + u" "),
voice_code=voice_code,
output_file_path=tmp_destination
)
self.log([u"Removing temporary file '%s'", tmp_destination])
gf.delete_file(handler, tmp_destination)
self.log(u"Synthesizing text... done")
return data
self.log(u"Calling TTS engine via subprocess...")
try:
# get sample rate and encoding
du_nu, sample_rate, encoding, da_nu = synthesize_and_clean(
text=u"Dummy text to get sample_rate",
voice_code=self._language_to_voice_code(self.DEFAULT_LANGUAGE)
)
# open output file
output_file = AudioFile(rconf=self.rconf, logger=self.logger)
output_file.audio_format = encoding
output_file.audio_channels = 1
output_file.audio_sample_rate = sample_rate
# create output
anchors = []
current_time = TimeValue("0.000")
num = 0
num_chars = 0
fragments = text_file.fragments
if backwards:
fragments = fragments[::-1]
for fragment in fragments:
# language to voice code
voice_code = self._language_to_voice_code(fragment.language)
# synthesize and get the duration of the output file
self.log([u"Synthesizing fragment %d", num])
duration, sr_nu, enc_nu, samples = synthesize_and_clean(
text=fragment.filtered_text,
voice_code=voice_code
)
# store for later output
anchors.append([current_time, fragment.identifier, fragment.text])
# increase the character counter
num_chars += fragment.characters
# concatenate new samples
self.log([u"Fragment %d starts at: %.3f", num, current_time])
if duration > 0:
self.log([u"Fragment %d duration: %.3f", num, duration])
current_time += duration
output_file.add_samples(samples, reverse=backwards)
else:
self.log([u"Fragment %d has zero duration", num])
# increment fragment counter
num += 1
# check if we must stop synthesizing because we have enough audio
if (quit_after is not None) and (current_time > quit_after):
self.log([u"Quitting after reached duration %.3f", current_time])
break
# minimize memory
self.log(u"Minimizing memory...")
output_file.minimize_memory()
self.log(u"Minimizing memory... done")
# if backwards, we need to reverse the audio samples again
if backwards:
self.log(u"Reversing audio samples...")
output_file.reverse()
self.log(u"Reversing audio samples... done")
# write output file
self.log([u"Writing audio file '%s'", output_file_path])
output_file.write(file_path=output_file_path)
except Exception as exc:
self.log_exc(u"An unexpected error occurred while calling TTS engine via subprocess", exc, False, None)
return (False, None)
# return output
if backwards:
self.log_warn(u"Please note that anchor time values do not make sense since backwards=True")
self.log([u"Returning %d time anchors", len(anchors)])
self.log([u"Current time %.3f", current_time])
self.log([u"Synthesized %d characters", num_chars])
self.log(u"Calling TTS engine via subprocess... done")
return (True, (anchors, current_time, num_chars))
def _synthesize_multiple_python(self,
text_file,
output_file_path,
quit_after=None,
backwards=False):
"""
Synthesize multiple text fragments, via Python call.
Return a tuple (anchors, total_time, num_chars).
:rtype: (bool, (list, TimeValue, int))
"""
#
# TODO in the Speect Python API I was not able to find a way
# to generate the wave incrementally
# so I essentially copy the subprocess call mechanism:
# generating wave data for each fragment,
# and concatenating them together
#
self.log(u"Calling TTS engine via Python...")
try:
# get sample rate and encoding
du_nu, sample_rate, encoding, da_nu = self._synthesize_single_helper(
text=u"Dummy text to get sample_rate",
voice_code=self.DEFAULT_LANGUAGE)
# open output file
output_file = AudioFile(rconf=self.rconf, logger=self.logger)
output_file.audio_format = encoding
output_file.audio_channels = 1
output_file.audio_sample_rate = sample_rate
# create output
anchors = []
current_time = TimeValue("0.000")
num = 0
num_chars = 0
fragments = text_file.fragments
if backwards:
fragments = fragments[::-1]
for fragment in fragments:
# language to voice code
#
# NOTE since voice_code is actually ignored
# in _synthesize_single_helper(),
# the value of voice_code is irrelevant
#
# however, in general you need to apply
# the _language_to_voice_code() function that maps
# the text language to a voice code
#
# here we apply the _language_to_voice_code() defined in super()
# that sets voice_code = fragment.language
#
voice_code = self._language_to_voice_code(fragment.language)
# synthesize and get the duration of the output file
self.log([u"Synthesizing fragment %d", num])
duration, sr_nu, enc_nu, data = self._synthesize_single_helper(
text=(fragment.filtered_text + u" "),
voice_code=voice_code)
# store for later output
anchors.append(
[current_time, fragment.identifier, fragment.text])
# increase the character counter
num_chars += fragment.characters
# append new data
self.log([u"Fragment %d starts at: %.3f", num, current_time])
if duration > 0:
self.log([u"Fragment %d duration: %.3f", num, duration])
current_time += duration
# if backwards, we append the data reversed
output_file.add_samples(data, reverse=backwards)
else:
self.log([u"Fragment %d has zero duration", num])
# increment fragment counter
num += 1
# check if we must stop synthesizing because we have enough audio
if (quit_after is not None) and (current_time > quit_after):
self.log([
u"Quitting after reached duration %.3f", current_time
])
break
# if backwards, we need to reverse the audio samples again
if backwards:
output_file.reverse()
# write output file
self.log([u"Writing audio file '%s'", output_file_path])
output_file.write(file_path=output_file_path)
except Exception as exc:
self.log_exc(
u"An unexpected error occurred while calling TTS engine via Python",
exc, False, None)
return (False, None)
# return output
# NOTE anchors do not make sense if backwards
self.log([u"Returning %d time anchors", len(anchors)])
self.log([u"Current time %.3f", current_time])
self.log([u"Synthesized %d characters", num_chars])
self.log(u"Calling TTS engine via Python... done")
return (True, (anchors, current_time, num_chars))
def _synthesize_multiple_python(self, text_file, output_file_path, quit_after=None, backwards=False):
"""
Synthesize multiple text fragments, via Python call.
Return a tuple (anchors, total_time, num_chars).
:rtype: (bool, (list, TimeValue, int))
"""
#
# TODO in the Speect Python API I was not able to find a way
# to generate the wave incrementally
# so I essentially copy the subprocess call mechanism:
# generating wave data for each fragment,
# and concatenating them together
#
self.log(u"Calling TTS engine via Python...")
try:
# get sample rate and encoding
du_nu, sample_rate, encoding, da_nu = self._synthesize_single_helper(
text=u"Dummy text to get sample_rate",
voice_code=self.DEFAULT_LANGUAGE
)
# open output file
output_file = AudioFile(rconf=self.rconf, logger=self.logger)
output_file.audio_format = encoding
output_file.audio_channels = 1
output_file.audio_sample_rate = sample_rate
# create output
anchors = []
current_time = TimeValue("0.000")
num = 0
num_chars = 0
fragments = text_file.fragments
if backwards:
fragments = fragments[::-1]
for fragment in fragments:
# language to voice code
#
# NOTE since voice_code is actually ignored
# in _synthesize_single_helper(),
# the value of voice_code is irrelevant
#
# however, in general you need to apply
# the _language_to_voice_code() function that maps
# the text language to a voice code
#
# here we apply the _language_to_voice_code() defined in super()
# that sets voice_code = fragment.language
#
voice_code = self._language_to_voice_code(fragment.language)
# synthesize and get the duration of the output file
self.log([u"Synthesizing fragment %d", num])
duration, sr_nu, enc_nu, data = self._synthesize_single_helper(
text=(fragment.filtered_text + u" "),
voice_code=voice_code
)
# store for later output
anchors.append([current_time, fragment.identifier, fragment.text])
# increase the character counter
num_chars += fragment.characters
# append new data
self.log([u"Fragment %d starts at: %.3f", num, current_time])
if duration > 0:
self.log([u"Fragment %d duration: %.3f", num, duration])
current_time += duration
# if backwards, we append the data reversed
output_file.add_samples(data, reverse=backwards)
else:
self.log([u"Fragment %d has zero duration", num])
# increment fragment counter
num += 1
# check if we must stop synthesizing because we have enough audio
if (quit_after is not None) and (current_time > quit_after):
self.log([u"Quitting after reached duration %.3f", current_time])
break
# if backwards, we need to reverse the audio samples again
if backwards:
output_file.reverse()
# write output file
self.log([u"Writing audio file '%s'", output_file_path])
output_file.write(file_path=output_file_path)
except Exception as exc:
self.log_exc(u"An unexpected error occurred while calling TTS engine via Python", exc, False, None)
return (False, None)
# return output
# NOTE anchors do not make sense if backwards
self.log([u"Returning %d time anchors", len(anchors)])
self.log([u"Current time %.3f", current_time])
self.log([u"Synthesized %d characters", num_chars])
self.log(u"Calling TTS engine via Python... done")
return (True, (anchors, current_time, num_chars))
def _cut_head_tail(self, audio_file_path):
"""
Set the audio file head or tail,
suitably cutting the audio file on disk,
and setting the corresponding parameters in the task configuration.
Return a success bool flag
"""
self._log("Setting head and/or tail")
try:
configuration = self.task.configuration
head_length = configuration.is_audio_file_head_length
process_length = configuration.is_audio_file_process_length
detect_head_min = configuration.is_audio_file_detect_head_min
detect_head_max = configuration.is_audio_file_detect_head_max
detect_tail_min = configuration.is_audio_file_detect_tail_min
detect_tail_max = configuration.is_audio_file_detect_tail_max
# explicit head or process?
explicit = (head_length is not None) or (process_length is not None)
# at least one detect parameter?
detect = (
(detect_head_min is not None) or
(detect_head_max is not None) or
(detect_tail_min is not None) or
(detect_tail_max is not None)
)
if explicit or detect:
# we need to load the audio data
audio_file = AudioFile(audio_file_path, logger=self.logger)
audio_file.load_data()
if explicit:
self._log("Explicit head or process")
else:
self._log("No explicit head or process => detecting head/tail")
head = 0.0
if (detect_head_min is not None) or (detect_head_max is not None):
self._log("Detecting head...")
detect_head_min = gf.safe_float(detect_head_min, gc.SD_MIN_HEAD_LENGTH)
detect_head_max = gf.safe_float(detect_head_max, gc.SD_MAX_HEAD_LENGTH)
self._log(["detect_head_min is %.3f", detect_head_min])
self._log(["detect_head_max is %.3f", detect_head_max])
sd = SD(audio_file, self.task.text_file, logger=self.logger)
head = sd.detect_head(detect_head_min, detect_head_max)
self._log(["Detected head: %.3f", head])
tail = 0.0
if (detect_tail_min is not None) or (detect_tail_max is not None):
self._log("Detecting tail...")
detect_tail_max = gf.safe_float(detect_tail_max, gc.SD_MAX_TAIL_LENGTH)
detect_tail_min = gf.safe_float(detect_tail_min, gc.SD_MIN_TAIL_LENGTH)
self._log(["detect_tail_min is %.3f", detect_tail_min])
self._log(["detect_tail_max is %.3f", detect_tail_max])
sd = SD(audio_file, self.task.text_file, logger=self.logger)
tail = sd.detect_tail(detect_tail_min, detect_tail_max)
self._log(["Detected tail: %.3f", tail])
# sanity check
head_length = max(0, head)
process_length = max(0, audio_file.audio_length - tail - head)
# we need to set these values
# in the config object for later use
self.task.configuration.is_audio_file_head_length = head_length
self.task.configuration.is_audio_file_process_length = process_length
self._log(["Set head_length: %.3f", head_length])
self._log(["Set process_length: %.3f", process_length])
if head_length is not None:
# in case we are reading from config object
head_length = float(head_length)
if process_length is not None:
# in case we are reading from config object
process_length = float(process_length)
# note that str() is necessary, as one might be None
self._log(["is_audio_file_head_length is %s", str(head_length)])
self._log(["is_audio_file_process_length is %s", str(process_length)])
self._log("Trimming audio data...")
audio_file.trim(head_length, process_length)
self._log("Trimming audio data... done")
self._log("Writing audio file...")
audio_file.write(audio_file_path)
self._log("Writing audio file... done")
audio_file.clear_data()
else:
# nothing to do
self._log("No explicit head/process or detect head/tail")
self._log("Setting head and/or tail: succeeded")
return True
except Exception as e:
self._log("Setting head and/or tail: failed")
self._log(["Message: %s", str(e)])
return False