GPT-SoVITS-v2-VC / tools /slice_audio.py
kevinwang676's picture
Upload folder using huggingface_hub
69cf514 verified
raw
history blame
2.18 kB
import os,sys,numpy as np
import traceback
from scipy.io import wavfile
# parent_directory = os.path.dirname(os.path.abspath(__file__))
# sys.path.append(parent_directory)
from tools.my_utils import load_audio
from slicer2 import Slicer
def slice(inp,opt_root,threshold,min_length,min_interval,hop_size,max_sil_kept,_max,alpha,i_part,all_part):
os.makedirs(opt_root,exist_ok=True)
if os.path.isfile(inp):
input=[inp]
elif os.path.isdir(inp):
input=[os.path.join(inp, name) for name in sorted(list(os.listdir(inp)))]
else:
return "输入路径存在但既不是文件也不是文件夹"
slicer = Slicer(
sr=32000, # 长音频采样率
threshold= int(threshold), # 音量小于这个值视作静音的备选切割点
min_length= int(min_length), # 每段最小多长,如果第一段太短一直和后面段连起来直到超过这个值
min_interval= int(min_interval), # 最短切割间隔
hop_size= int(hop_size), # 怎么算音量曲线,越小精度越大计算量越高(不是精度越大效果越好)
max_sil_kept= int(max_sil_kept), # 切完后静音最多留多长
)
_max=float(_max)
alpha=float(alpha)
for inp_path in input[int(i_part)::int(all_part)]:
# print(inp_path)
try:
name = os.path.basename(inp_path)
audio = load_audio(inp_path, 32000)
# print(audio.shape)
for chunk, start, end in slicer.slice(audio): # start和end是帧数
tmp_max = np.abs(chunk).max()
if(tmp_max>1):chunk/=tmp_max
chunk = (chunk / tmp_max * (_max * alpha)) + (1 - alpha) * chunk
wavfile.write(
"%s/%s_%010d_%010d.wav" % (opt_root, name, start, end),
32000,
# chunk.astype(np.float32),
(chunk * 32767).astype(np.int16),
)
except:
print(inp_path,"->fail->",traceback.format_exc())
return "执行完毕,请检查输出文件"
print(slice(*sys.argv[1:]))