storytold/storyteller-ml
2
0
Fork 0
mirror of https://github.com/storytold/storyteller-ml.git synced 2026-10-26 08:48:03 +00:00
Code Issues Packages Projects Releases Wiki Activity

support DIO

Browse Source
This commit is contained in:
Brandon Thomas
2023-10-10 01:07:51 -04:00
parent b0856e9426
commit b41538fa61
1 changed files with 17 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files
voice_conversion/rvc-v2-brandon/vc_infer_pipeline.py
+17
View File
@@ -27,6 +27,18 @@ def cache_harvest_f0(input_audio_path, fs, f0max, f0min, frame_period):
f0 = pyworld.stonemask(audio, f0, t, fs) f0 = pyworld.stonemask(audio, f0, t, fs)
return f0 return f0
@lru_cache
def cache_dio_f0(input_audio_path, fs, f0max, f0min, frame_period):
audio = input_audio_path2wav[input_audio_path]
f0, t = pyworld.dio(
audio,
fs=fs,
f0_ceil=f0max,
f0_floor=f0min,
frame_period=frame_period,
)
f0 = pyworld.stonemask(audio, f0, t, fs)
return f0
def change_rms(data1, sr1, data2, sr2, rate): # 1是输入音频，2是输出音频,rate是2的占比 def change_rms(data1, sr1, data2, sr2, rate): # 1是输入音频，2是输出音频,rate是2的占比
# print(data1.max(),data2.max()) # print(data1.max(),data2.max())
@@ -107,6 +119,11 @@ class VC(object):
f0 = cache_harvest_f0(input_audio_path, self.sr, f0_max, f0_min, 10) f0 = cache_harvest_f0(input_audio_path, self.sr, f0_max, f0_min, 10)
if filter_radius > 2: if filter_radius > 2:
f0 = signal.medfilt(f0, 3) f0 = signal.medfilt(f0, 3)
elif f0_method == "dio":
input_audio_path2wav[input_audio_path] = x.astype(np.double)
f0 = cache_dio_f0(input_audio_path, self.sr, f0_max, f0_min, 10)
if filter_radius > 2:
f0 = signal.medfilt(f0, 3)
elif f0_method == "crepe": elif f0_method == "crepe":
model = "full" model = "full"
# Pick a batch size that doesn't cause memory errors on your gpu # Pick a batch size that doesn't cause memory errors on your gpu