fumiama/Retrieval-based-Voice-Conversion-WebUI
1
0
Fork 0
mirror of https://github.com/fumiama/Retrieval-based-Voice-Conversion-WebUI.git synced 2026-06-07 10:10:23 +08:00
Code Activity

optimize: some training optimizations (#95)

Browse Source 
* optimzie(train&uvr5): rm sf & simp. AudioPre

* fix(audio): too many mallocs

* feat(audio): load_audio support stereo

* fix(audio): float32 wav saving

* fix(train): missing ckpt var
This commit is contained in:
源文雨
2024-11-28 03:20:14 +09:00
committed by GitHub
parent f4644ec1ec
commit a8783c6639
19 changed files with 163 additions and 433 deletions
Download Patch File Download Diff File Expand all files Collapse all files

18
infer/modules/vc/modules.py
View File

@@ -5,11 +5,10 @@ import os
logger = logging.getLogger(__name__)
import numpy as np
import soundfile as sf
import torch
from io import BytesIO
from infer.lib.audio import load_audio, wav2
from infer.lib.audio import load_audio, wav2, save_audio, float_np_array_to_wav_buf
from rvc.synthesizer import get_synthesizer, load_synthesizer
from .info import show_model_info
from .pipeline import Pipeline
@@ -253,23 +252,16 @@ class VC:
try:
tgt_sr, audio_opt = opt
if format1 in ["wav", "flac"]:
sf.write(
"%s/%s.%s"
% (opt_root, os.path.basename(path), format1),
audio_opt,
tgt_sr,
)
save_audio("%s/%s.%s"
% (opt_root, os.path.basename(path), format1), audio_opt, tgt_sr)
else:
path = "%s/%s.%s" % (
opt_root,
os.path.basename(path),
format1,
)
with BytesIO() as wavf:
sf.write(wavf, audio_opt, tgt_sr, format="wav")
wavf.seek(0, 0)
with open(path, "wb") as outf:
wav2(wavf, outf, format1)
with open(path, "wb") as outf:
wav2(float_np_array_to_wav_buf(audio_opt, tgt_sr), outf, format1)
except:
info += traceback.format_exc()
infos.append("%s->%s" % (os.path.basename(path), info))