bacow/GPT-SoVITS
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Merge branch 'main' into main

Browse Source
This commit is contained in:
RVC-Boss
2024-01-29 23:25:48 +08:00
committed by GitHub
parent 21c0ab1233 5afe04cf3a
commit d98dd2dd26
51 changed files with 5441 additions and 1101 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
tools/my_utils.py
View File

@@ -1,4 +1,4 @@
import platform
import platform,os,traceback
import ffmpeg
import numpy as np
@@ -9,12 +9,17 @@ def load_audio(file, sr):
# This launches a subprocess to decode audio while down-mixing and resampling as necessary.
# Requires the ffmpeg CLI and `ffmpeg-python` package to be installed.
file = clean_path(file) # 防止小白拷路径头尾带了空格和"和回车
if os.path.exists(file) == False:
raise RuntimeError(
"You input a wrong audio path that does not exists, please fix it!"
)
out, _ = (
ffmpeg.input(file, threads=0)
.output("-", format="f32le", acodec="pcm_f32le", ac=1, ar=sr)
.run(cmd=["ffmpeg", "-nostdin"], capture_stdout=True, capture_stderr=True)
)
except Exception as e:
traceback.print_exc()
raise RuntimeError(f"Failed to load audio: {e}")
return np.frombuffer(out, np.float32).flatten()

6
tools/subfix_webui.py
View File

@@ -110,6 +110,7 @@ def b_submit_change(*text_list):
def b_delete_audio(*checkbox_list):
global g_data_json, g_index, g_max_json_index
b_save_file()
change = False
for i, checkbox in reversed(list(enumerate(checkbox_list))):
if g_index + i < len(g_data_json):
@@ -121,8 +122,8 @@ def b_delete_audio(*checkbox_list):
if g_index > g_max_json_index:
g_index = g_max_json_index
g_index = g_index if g_index >= 0 else 0
# if change:
# b_save_file()
if change:
b_save_file()
# return gr.Slider(value=g_index, maximum=(g_max_json_index if g_max_json_index>=0 else 0)), *b_change_index(g_index, g_batch)
return {"value":g_index,"__type__":"update","maximum":(g_max_json_index if g_max_json_index>=0 else 0)},*b_change_index(g_index, g_batch)
@@ -172,6 +173,7 @@ def b_audio_split(audio_breakpoint, *checkbox_list):
def b_merge_audio(interval_r, *checkbox_list):
global g_data_json , g_max_json_index
b_save_file()
checked_index = []
audios_path = []
audios_text = []

3
tools/uvr5/vr.py
View File

@@ -110,6 +110,9 @@ class AudioPre:
y_spec_m = pred * X_phase
v_spec_m = X_spec_m - y_spec_m
if is_hp3 == True:
ins_root,vocal_root = vocal_root,ins_root
if ins_root is not None:
if self.data["high_end_process"].startswith("mirroring"):
input_high_end_ = spec_utils.mirroring(

88
tools/uvr5/webui.py
View File

@@ -5,7 +5,8 @@ from tools.i18n.i18n import I18nAuto
i18n = I18nAuto()
logger = logging.getLogger(__name__)
import ffmpeg
import librosa
import soundfile as sf
import torch
import sys
from mdxnet import MDXNetDereverb
@@ -51,52 +52,55 @@ def uvr(model_name, inp_root, save_root_vocal, paths, save_root_ins, agg, format
paths = [path.name for path in paths]
for path in paths:
inp_path = os.path.join(inp_root, path)
need_reformat = 1
done = 0
if(os.path.isfile(inp_path)==False):continue
try:
info = ffmpeg.probe(inp_path, cmd="ffprobe")
if (
info["streams"][0]["channels"] == 2
and info["streams"][0]["sample_rate"] == "44100"
):
need_reformat = 0
pre_fun._path_audio_(
inp_path, save_root_ins, save_root_vocal, format0, is_hp3=is_hp3
done = 0
try:
y, sr = librosa.load(inp_path, sr=None)
info = sf.info(inp_path)
channels = info.channels
if channels == 2 and sr == 44100:
need_reformat = 0
pre_fun._path_audio_(
inp_path, save_root_ins, save_root_vocal, format0, is_hp3=is_hp3
)
done = 1
else:
need_reformat = 1
except:
need_reformat = 1
traceback.print_exc()
if need_reformat == 1:
tmp_path = "%s/%s.reformatted.wav" % (
os.path.join(os.environ["TEMP"]),
os.path.basename(inp_path),
)
done = 1
except:
need_reformat = 1
traceback.print_exc()
if need_reformat == 1:
tmp_path = "%s/%s.reformatted.wav" % (
os.path.join(os.environ["TEMP"]),
os.path.basename(inp_path),
)
os.system(
"ffmpeg -i %s -vn -acodec pcm_s16le -ac 2 -ar 44100 %s -y"
% (inp_path, tmp_path)
)
inp_path = tmp_path
try:
if done == 0:
pre_fun._path_audio_(
inp_path, save_root_ins, save_root_vocal, format0
)
infos.append("%s->Success" % (os.path.basename(inp_path)))
yield "\n".join(infos)
except:
y_resampled = librosa.resample(y, sr, 44100)
sf.write(tmp_path, y_resampled, 44100, "PCM_16")
inp_path = tmp_path
try:
if done == 0:
pre_fun._path_audio_(
inp_path, save_root_ins, save_root_vocal, format0
inp_path, save_root_ins, save_root_vocal, format0, is_hp3=is_hp3
)
infos.append("%s->Success" % (os.path.basename(inp_path)))
yield "\n".join(infos)
except:
infos.append(
"%s->%s" % (os.path.basename(inp_path), traceback.format_exc())
)
yield "\n".join(infos)
try:
if done == 0:
pre_fun._path_audio_(
inp_path, save_root_ins, save_root_vocal, format0, is_hp3=is_hp3
)
infos.append("%s->Success" % (os.path.basename(inp_path)))
yield "\n".join(infos)
except:
infos.append(
"%s->%s" % (os.path.basename(inp_path), traceback.format_exc())
)
yield "\n".join(infos)
except:
infos.append("Oh my god. %s->%s"%(os.path.basename(inp_path), traceback.format_exc()))
yield "\n".join(infos)
except:
infos.append(traceback.format_exc())
yield "\n".join(infos)
@@ -116,10 +120,10 @@ def uvr(model_name, inp_root, save_root_vocal, paths, save_root_ins, agg, format
yield "\n".join(infos)
with gr.Blocks(title="RVC WebUI") as app:
with gr.Blocks(title="UVR5 WebUI") as app:
gr.Markdown(
value=
"本软件以MIT协议开源, 作者不对软件具备任何控制力, 使用软件者、传播软件导出的声音者自负全责. <br>如不认可该条款, 则不能使用或引用软件包内任何代码和文件. 详见根目录<b>LICENSE</b>."
i18n("本软件以MIT协议开源, 作者不对软件具备任何控制力, 使用软件者、传播软件导出的声音者自负全责. <br>如不认可该条款, 则不能使用或引用软件包内任何代码和文件. 详见根目录<b>LICENSE</b>.")
)
with gr.Tabs():
with gr.TabItem(i18n("伴奏人声分离&去混响&去回声")):
@@ -144,7 +148,7 @@ with gr.Blocks(title="RVC WebUI") as app:
minimum=0,
maximum=20,
step=1,
label="人声提取激进程度",
label=i18n("人声提取激进程度"),
value=10,
interactive=True,
visible=False, # 先不开放调整
@@ -183,4 +187,4 @@ app.queue(concurrency_count=511, max_size=1022).launch(
share=is_share,
server_port=webui_port_uvr5,
quiet=True,
)
)