fumiama/Retrieval-based-Voice-Conversion-WebUI
1
0
Fork 0
mirror of https://github.com/fumiama/Retrieval-based-Voice-Conversion-WebUI.git synced 2026-06-06 09:40:24 +08:00
Code Activity

chore(format): run black on dev (#94)

Browse Source 
Co-authored-by: github-actions[bot] <github-actions[bot]@users.noreply.github.com>
This commit is contained in:
github-actions[bot]
2024-11-28 03:21:10 +09:00
committed by GitHub
parent a8783c6639
commit d3add81469
10 changed files with 126 additions and 47 deletions
Download Patch File Download Diff File Expand all files Collapse all files

25
infer/modules/train/preprocess.py
View File

@@ -62,15 +62,24 @@ class PreProcess:
tmp_audio = (tmp_audio / tmp_max * (self.max * self.alpha)) + (
1 - self.alpha
) * tmp_audio
save_audio("%s/%s_%s.wav" % (self.gt_wavs_dir, idx0, idx1), tmp_audio, self.sr, f32=True)
save_audio(
"%s/%s_%s.wav" % (self.gt_wavs_dir, idx0, idx1),
tmp_audio,
self.sr,
f32=True,
)
with open("%s/%s_%s.wav" % (self.wavs16k_dir, idx0, idx1), "wb") as f:
f.write(float_np_array_to_wav_buf(
load_audio(
float_np_array_to_wav_buf(tmp_audio, self.sr, f32=True),
sr=16000,
format="wav",
)
, 16000, True).getbuffer())
f.write(
float_np_array_to_wav_buf(
load_audio(
float_np_array_to_wav_buf(tmp_audio, self.sr, f32=True),
sr=16000,
format="wav",
),
16000,
True,
).getbuffer()
)
def pipeline(self, path, idx0):
try:

30
infer/modules/train/train.py
View File

@@ -133,11 +133,21 @@ def run(rank, n_gpus, hps: utils.HParams, logger: logging.Logger):
try:
dist.init_process_group(
backend="gloo" if os.name == "nt" or not torch.cuda.is_available() else "nccl", init_method="env://", world_size=n_gpus, rank=rank
backend=(
"gloo" if os.name == "nt" or not torch.cuda.is_available() else "nccl"
),
init_method="env://",
world_size=n_gpus,
rank=rank,
)
except:
dist.init_process_group(
backend="gloo" if os.name == "nt" or not torch.cuda.is_available() else "nccl", init_method="env://?use_libuv=False", world_size=n_gpus, rank=rank
backend=(
"gloo" if os.name == "nt" or not torch.cuda.is_available() else "nccl"
),
init_method="env://?use_libuv=False",
world_size=n_gpus,
rank=rank,
)
torch.manual_seed(hps.train.seed)
if torch.cuda.is_available():
@@ -243,13 +253,17 @@ def run(rank, n_gpus, hps: utils.HParams, logger: logging.Logger):
if hasattr(net_g, "module"):
logger.info(
net_g.module.load_state_dict(
torch.load(hps.pretrainG, map_location="cpu", weights_only=True)["model"]
torch.load(
hps.pretrainG, map_location="cpu", weights_only=True
)["model"]
)
) ##测试不加载优化器
else:
logger.info(
net_g.load_state_dict(
torch.load(hps.pretrainG, map_location="cpu", weights_only=True)["model"]
torch.load(
hps.pretrainG, map_location="cpu", weights_only=True
)["model"]
)
) ##测试不加载优化器
if hps.pretrainD != "":
@@ -258,13 +272,17 @@ def run(rank, n_gpus, hps: utils.HParams, logger: logging.Logger):
if hasattr(net_d, "module"):
logger.info(
net_d.module.load_state_dict(
torch.load(hps.pretrainD, map_location="cpu", weights_only=True)["model"]
torch.load(
hps.pretrainD, map_location="cpu", weights_only=True
)["model"]
)
)
else:
logger.info(
net_d.load_state_dict(
torch.load(hps.pretrainD, map_location="cpu", weights_only=True)["model"]
torch.load(
hps.pretrainD, map_location="cpu", weights_only=True
)["model"]
)
)

8
infer/modules/uvr5/mdxnet.py
View File

@@ -208,8 +208,12 @@ class Predictor:
sources = self.demix(mix.T)
opt = sources[0].T
if format in ["wav", "flac"]:
save_audio("%s/vocal_%s.%s" % (vocal_root, basename, format), mix - opt, rate)
save_audio("%s/instrument_%s.%s" % (others_root, basename, format), opt, rate)
save_audio(
"%s/vocal_%s.%s" % (vocal_root, basename, format), mix - opt, rate
)
save_audio(
"%s/instrument_%s.%s" % (others_root, basename, format), opt, rate
)
else:
path_vocal = "%s/vocal_%s.wav" % (vocal_root, basename)
path_other = "%s/instrument_%s.wav" % (others_root, basename)

20
infer/modules/uvr5/vr.py
View File

@@ -48,9 +48,7 @@ class AudioPre:
self.mp = mp
self.model = model
def _path_audio_(
self, music_file, ins_root=None, vocal_root=None, format="flac"
):
def _path_audio_(self, music_file, ins_root=None, vocal_root=None, format="flac"):
if ins_root is None and vocal_root is None:
return "No save root."
name = os.path.basename(music_file)
@@ -134,10 +132,14 @@ class AudioPre:
else:
head = "instrument_"
if format in ["wav", "flac"]:
save_audio(os.path.join(
save_audio(
os.path.join(
ins_root,
head + "{}_{}.{}".format(name, self.data["agg"], format),
), wav_instrument, self.mp.param["sr"])
),
wav_instrument,
self.mp.param["sr"],
)
else:
path = os.path.join(
ins_root, head + "{}_{}.wav".format(name, self.data["agg"])
@@ -162,10 +164,14 @@ class AudioPre:
wav_vocals = spec_utils.cmb_spectrogram_to_wave(v_spec_m, self.mp)
logger.info("%s vocals done" % name)
if format in ["wav", "flac"]:
save_audio(os.path.join(
save_audio(
os.path.join(
vocal_root,
head + "{}_{}.{}".format(name, self.data["agg"], format),
), wav_vocals, self.mp.param["sr"])
),
wav_vocals,
self.mp.param["sr"],
)
else:
path = os.path.join(
vocal_root, head + "{}_{}.wav".format(name, self.data["agg"])

14
infer/modules/vc/modules.py
View File

@@ -252,8 +252,12 @@ class VC:
try:
tgt_sr, audio_opt = opt
if format1 in ["wav", "flac"]:
save_audio("%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,
@@ -261,7 +265,11 @@ class VC:
format1,
)
with open(path, "wb") as outf:
wav2(float_np_array_to_wav_buf(audio_opt, tgt_sr), outf, format1)
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))