RVC/Retrieval-based-Voice-Conversion-WebUI
1
0
Fork 0
mirror of https://github.com/RVC-Project/Retrieval-based-Voice-Conversion-WebUI.git synced 2026-02-20 19:30:49 +00:00
Code Issues Packages Projects Releases Wiki Activity

768VecOnnxExport (#328)

Browse Source 
* Delete export_onnx.py

* Delete export_onnx_old.py

* Delete models_onnx_moess.py

* Support 768 Vec

* Add files via upload

* Support 768 Vec

Support 768 Vec

* Support 768 Vec Onnx Export

Support 768 Vec Onnx Export
This commit is contained in:
Ναρουσέ·μ·γιουμεμί·Χινακάννα
2023-05-21 19:11:29 +08:00
committed by GitHub
parent c3de24f2e0
commit 067731db9b
5 changed files with 144 additions and 1046 deletions
Download Patch File Download Diff File Expand all files Collapse all files

98
export_onnx.py
View File

@@ -1,5 +1,4 @@
from infer_pack.models_onnx_moess import SynthesizerTrnMs256NSFsidM
from infer_pack.models_onnx import SynthesizerTrnMs256NSFsidO
from infer_pack.models_onnx import SynthesizerTrnMsNSFsidM
import torch
if __name__ == "__main__":
@@ -21,65 +20,36 @@ if __name__ == "__main__":
device = "cpu" # 导出时设备（不影响使用模型）
if MoeVS:
net_g = SynthesizerTrnMs256NSFsidM(
*cpt["config"], is_half=False
) # fp32导出（C++要支持fp16必须手动将内存重新排列所以暂时不用fp16）
net_g.load_state_dict(cpt["weight"], strict=False)
input_names = ["phone", "phone_lengths", "pitch", "pitchf", "ds", "rnd"]
output_names = [
"audio",
]
torch.onnx.export(
net_g,
(
test_phone.to(device),
test_phone_lengths.to(device),
test_pitch.to(device),
test_pitchf.to(device),
test_ds.to(device),
test_rnd.to(device),
),
ExportedPath,
dynamic_axes={
"phone": [1],
"pitch": [1],
"pitchf": [1],
"rnd": [2],
},
do_constant_folding=False,
opset_version=16,
verbose=False,
input_names=input_names,
output_names=output_names,
)
else:
net_g = SynthesizerTrnMs256NSFsidO(
*cpt["config"], is_half=False
) # fp32导出（C++要支持fp16必须手动将内存重新排列所以暂时不用fp16）
net_g.load_state_dict(cpt["weight"], strict=False)
input_names = ["phone", "phone_lengths", "pitch", "pitchf", "ds"]
output_names = [
"audio",
]
torch.onnx.export(
net_g,
(
test_phone.to(device),
test_phone_lengths.to(device),
test_pitch.to(device),
test_pitchf.to(device),
test_ds.to(device),
),
ExportedPath,
dynamic_axes={
"phone": [1],
"pitch": [1],
"pitchf": [1],
},
do_constant_folding=False,
opset_version=16,
verbose=False,
input_names=input_names,
output_names=output_names,
)
net_g = SynthesizerTrnMsNSFsidM(
*cpt["config"], is_half=False
) # fp32导出（C++要支持fp16必须手动将内存重新排列所以暂时不用fp16）
net_g.load_state_dict(cpt["weight"], strict=False)
input_names = ["phone", "phone_lengths", "pitch", "pitchf", "ds", "rnd"]
output_names = [
"audio",
]
# net_g.construct_spkmixmap(n_speaker) 多角色混合轨道导出
torch.onnx.export(
net_g,
(
test_phone.to(device),
test_phone_lengths.to(device),
test_pitch.to(device),
test_pitchf.to(device),
test_ds.to(device),
test_rnd.to(device),
),
ExportedPath,
dynamic_axes={
"phone": [1],
"pitch": [1],
"pitchf": [1],
"rnd": [2],
},
do_constant_folding=False,
opset_version=16,
verbose=False,
input_names=input_names,
output_names=output_names,
)