Medix-AI/coqui-tts
Medix-AI
/
coqui-tts
mirror of https://github.com/coqui-ai/TTS.git
1
0
Fork 0
Code Issues Packages Projects Releases Wiki Activity

unpin cython version and commentout pyworld in audio.py causing dep 

issues
This commit is contained in:
Eren Gölge 2021-02-01 11:34:05 +00:00
parent 5beed0ddcd
commit 8774e37444
2 changed files with 12 additions and 11 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

21
TTS/utils/audio.py
View File

@ -3,7 +3,7 @@ import soundfile as sf
import numpy as np import numpy as np
import scipy.io.wavfile import scipy.io.wavfile
import scipy.signal import scipy.signal
import pyworld as pw # import pyworld as pw
from TTS.tts.utils.data import StandardScaler from TTS.tts.utils.data import StandardScaler
@ -292,15 +292,16 @@ class AudioProcessor(object):
return pad // 2, pad // 2 + pad % 2 return pad // 2, pad // 2 + pad % 2
### Compute F0 ### ### Compute F0 ###
def compute_f0(self, x): # TODO: pw causes some dep issues
f0, t = pw.dio( # def compute_f0(self, x):
x.astype(np.double), # f0, t = pw.dio(
fs=self.sample_rate, # x.astype(np.double),
f0_ceil=self.mel_fmax, # fs=self.sample_rate,
frame_period=1000 * self.hop_length / self.sample_rate, # f0_ceil=self.mel_fmax,
) # frame_period=1000 * self.hop_length / self.sample_rate,
f0 = pw.stonemask(x.astype(np.double), f0, t, self.sample_rate) # )
return f0 # f0 = pw.stonemask(x.astype(np.double), f0, t, self.sample_rate)
# return f0
### Audio Processing ### ### Audio Processing ###
def find_endpoint(self, wav, threshold_db=-40, min_silence_sec=0.8): def find_endpoint(self, wav, threshold_db=-40, min_silence_sec=0.8):

2
requirements.txt
View File

@ -21,5 +21,5 @@ cardboardlint==1.3.0
pylint==2.5.3 pylint==2.5.3
gdown gdown
umap-learn umap-learn
cython==0.29.20 # > 0.29.20 breaks pyworld installation with the min numpy req of Tensorflow 2.4.1 cython
pyyaml pyyaml