backup/Realtime-Voice-Clone-Chinese
1
0
Fork 0
mirror of https://github.com/babysor/Realtime-Voice-Clone-Chinese.git synced 2026-04-28 12:50:52 +08:00
Code Issues Projects Releases Wiki Activity

Some changes to make it easier to install the dependencies

Browse Source
This commit is contained in:
0warning0error
2023-06-02 17:22:38 +08:00
parent b78d0d2a26
commit 9f1dbeeecc
13 changed files with 93 additions and 46 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
models/encoder/audio.py
View File

@@ -39,7 +39,7 @@ def preprocess_wav(fpath_or_wav: Union[str, Path, np.ndarray],
# Resample the wav if needed
if source_sr is not None and source_sr != sampling_rate:
wav = librosa.resample(wav, source_sr, sampling_rate)
wav = librosa.resample(wav, orig_sr = source_sr, target_sr = sampling_rate)
# Apply the preprocessing: normalize volume and shorten long silences
if normalize:
@@ -99,7 +99,7 @@ def trim_long_silences(wav):
return ret[width - 1:] / width
audio_mask = moving_average(voice_flags, vad_moving_average_width)
audio_mask = np.round(audio_mask).astype(np.bool)
audio_mask = np.round(audio_mask).astype(bool)
# Dilate the voiced regions
audio_mask = binary_dilation(audio_mask, np.ones(vad_max_silence_length + 1))

2
models/encoder/visualizations.py
View File

@@ -21,7 +21,7 @@ colormap = np.array([
[33, 0, 127],
[0, 0, 0],
[183, 183, 183],
], dtype=np.float) / 255
], dtype=float) / 255
class Visualizations: