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

refactor(audio.processor): use trim_silence from numpy_transforms

This commit is contained in:
Enno Hermann 2023-11-14 13:39:37 +01:00
parent 842a632cd5
commit 0a0e7a3bae
1 changed files with 8 additions and 5 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

13
TTS/utils/audio/processor.py
View File

@ -22,6 +22,7 @@ from TTS.utils.audio.numpy_transforms import (
rms_volume_norm, rms_volume_norm,
spec_to_mel, spec_to_mel,
stft, stft,
trim_silence,
) )
# pylint: disable=too-many-public-methods # pylint: disable=too-many-public-methods
@ -539,11 +540,13 @@ class AudioProcessor(object):
def trim_silence(self, wav): def trim_silence(self, wav):
"""Trim silent parts with a threshold and 0.01 sec margin""" """Trim silent parts with a threshold and 0.01 sec margin"""
margin = int(self.sample_rate * 0.01) return trim_silence(
wav = wav[margin:-margin] wav=wav,
return librosa.effects.trim(wav, top_db=self.trim_db, frame_length=self.win_length, hop_length=self.hop_length)[ sample_rate=self.sample_rate,
0 trim_db=self.trim_db,
] win_length=self.win_length,
hop_length=self.hop_length,
)
@staticmethod @staticmethod
def sound_norm(x: np.ndarray) -> np.ndarray: def sound_norm(x: np.ndarray) -> np.ndarray: