skeh/ovtk_audiencekit
1
0
Fork 0
Code Issues 1 Pull requests Projects Releases Packages Wiki Activity

Compare commits

base: skeh:2d908e60a52847b23ab3a7b6276dc70111db5082
Branches Tags
skeh:main
skeh:feat/windows
skeh:wip/modqueue
skeh:feat/revitalize
skeh:feat/better-yt
skeh:plugin-based-chats
skeh:feat/tts
skeh:feat/mkeventFactory
..
compare: skeh:11b4c92fe986b10a7e2e1bd5dfa0c3aaf50378f4
Branches Tags
skeh:feat/windows
skeh:main
skeh:wip/modqueue
skeh:feat/revitalize
skeh:feat/better-yt
skeh:plugin-based-chats
skeh:feat/tts
skeh:feat/mkeventFactory

No commits in common. "2d908e60a52847b23ab3a7b6276dc70111db5082" and "11b4c92fe986b10a7e2e1bd5dfa0c3aaf50378f4" have entirely different histories.
2d908e60a5 ... 11b4c92fe9

2 changed files with 1 additions and 3 deletions
Show stats Expand all files Collapse all files

2
src/ovtk_audiencekit/core/Audio.py
View file

@ -53,7 +53,7 @@ class Clip:
def stretch(self, speed, keep_pitch=True): def stretch(self, speed, keep_pitch=True):
if keep_pitch: if keep_pitch:
stretched = tsm.wsola(self._stereo_transpose(self.raw), 1 / speed) stretched = tsm.wsola(self._stereo_transpose(self.raw), speed)
else: else:
stretched = librosa.resample(self._stereo_transpose(self.raw), self.samplerate * speed, self.samplerate, fix=False, scale=True) stretched = librosa.resample(self._stereo_transpose(self.raw), self.samplerate * speed, self.samplerate, fix=False, scale=True)
self.raw = np.ascontiguousarray(self._stereo_transpose(stretched), dtype='float32') self.raw = np.ascontiguousarray(self._stereo_transpose(stretched), dtype='float32')

2
src/ovtk_audiencekit/plugins/TTS/TTS.py
View file

@ -74,14 +74,12 @@ class TextToSpeechPlugin(PluginBase):
if filename is None: if filename is None:
filename = os.path.join(self.cache_dir, f'{uuid.uuid1()}.wav') filename = os.path.join(self.cache_dir, f'{uuid.uuid1()}.wav')
self.logger.info(f'Generating TTS "{text}"...')
if self.speaker_wav: if self.speaker_wav:
wav = self.synthesizer.tts(text, None, 'en', self.speaker_wav) wav = self.synthesizer.tts(text, None, 'en', self.speaker_wav)
else: else:
wav = self.synthesizer.tts(text) wav = self.synthesizer.tts(text)
self.synthesizer.save_wav(wav, filename) self.synthesizer.save_wav(wav, filename)
self.logger.info(f'Done - saved as {filename}')
return filename return filename
async def run(self, text, *args, _ctx={}, wait=False, **kwargs): async def run(self, text, *args, _ctx={}, wait=False, **kwargs):