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

resample audio if flag set true

This commit is contained in:
erogol 2020-11-26 11:30:30 +01:00
parent f6c96b0ac2
commit 0c6f7e4c77
1 changed files with 4 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

4
TTS/utils/audio.py
View File

@ -11,6 +11,7 @@ from TTS.tts.utils.data import StandardScaler
class AudioProcessor(object): class AudioProcessor(object):
def __init__(self, def __init__(self,
sample_rate=None, sample_rate=None,
resample=False,
num_mels=None, num_mels=None,
min_level_db=None, min_level_db=None,
frame_shift_ms=None, frame_shift_ms=None,
@ -39,6 +40,7 @@ class AudioProcessor(object):
print(" > Setting up Audio Processor...") print(" > Setting up Audio Processor...")
# setup class attributed # setup class attributed
self.sample_rate = sample_rate self.sample_rate = sample_rate
self.resample = resample
self.num_mels = num_mels self.num_mels = num_mels
self.min_level_db = min_level_db or 0 self.min_level_db = min_level_db or 0
self.frame_shift_ms = frame_shift_ms self.frame_shift_ms = frame_shift_ms
@ -324,6 +326,8 @@ class AudioProcessor(object):
if sr is None: if sr is None:
x, sr = sf.read(filename) x, sr = sf.read(filename)
assert self.sample_rate == sr, "%s vs %s"%(self.sample_rate, sr) assert self.sample_rate == sr, "%s vs %s"%(self.sample_rate, sr)
elif self.resample:
x, sr = librosa.load(filename, sr=self.sample_rate)
else: else:
x, sr = librosa.load(filename, sr=sr) x, sr = librosa.load(filename, sr=sr)
if self.do_trim_silence: if self.do_trim_silence: