so-vits-svc/onnx_export.py

103 lines
3.9 KiB
Python

import torch
from torchaudio.models.wav2vec2.utils import import_fairseq_model
from fairseq import checkpoint_utils
from onnxexport.model_onnx import SynthesizerTrn
import utils
def get_hubert_model():
vec_path = "hubert/checkpoint_best_legacy_500.pt"
print("load model(s) from {}".format(vec_path))
models, saved_cfg, task = checkpoint_utils.load_model_ensemble_and_task(
[vec_path],
suffix="",
)
model = models[0]
model.eval()
return model
def main(HubertExport, NetExport):
path = "SoVits4.0V2"
'''if HubertExport:
device = torch.device("cpu")
vec_path = "hubert/checkpoint_best_legacy_500.pt"
models, saved_cfg, task = checkpoint_utils.load_model_ensemble_and_task(
[vec_path],
suffix="",
)
original = models[0]
original.eval()
model = original
test_input = torch.rand(1, 1, 16000)
model(test_input)
torch.onnx.export(model,
test_input,
"hubert4.0.onnx",
export_params=True,
opset_version=16,
do_constant_folding=True,
input_names=['source'],
output_names=['embed'],
dynamic_axes={
'source':
{
2: "sample_length"
},
}
)'''
if NetExport:
device = torch.device("cpu")
hps = utils.get_hparams_from_file(f"checkpoints/{path}/config.json")
SVCVITS = SynthesizerTrn(
hps)
_ = utils.load_checkpoint(f"checkpoints/{path}/model.pth", SVCVITS, None)
_ = SVCVITS.eval().to(device)
for i in SVCVITS.parameters():
i.requires_grad = False
test_hidden_unit = torch.rand(1, 10, 256) # rand
test_pitch = torch.rand(1, 10) # rand
test_mel2ph = torch.LongTensor([0, 1, 2, 3, 4, 5, 6, 7, 8, 9]).unsqueeze(0)
#test_mel2ph = torch.LongTensor([0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19]).unsqueeze(0)
test_uv = torch.zeros(1, 2048, 10, dtype=torch.float32)
test_uv += 0.0064
test_noise = torch.randn(1, 192, 10) # randn
test_sid = torch.LongTensor([0])
input_names = ["c", "f0", "mel2ph", "t_window", "noise", "sid"]
output_names = ["audio", ]
SVCVITS.eval()
SVCVITS(test_hidden_unit.to(device),
test_pitch.to(device),
test_mel2ph.to(device),
test_uv.to(device),
test_noise.to(device),
test_sid.to(device))
torch.onnx.export(SVCVITS,
(
test_hidden_unit.to(device),
test_pitch.to(device),
test_mel2ph.to(device),
test_uv.to(device),
test_noise.to(device),
test_sid.to(device)
),
f"checkpoints/{path}/model.onnx",
dynamic_axes={
"c": [0, 1],
"f0": [1],
"mel2ph": [1],
"t_window": [2],
"noise": [2],
},
do_constant_folding=False,
opset_version=16,
verbose=False,
input_names=input_names,
output_names=output_names)
if __name__ == '__main__':
main(False, True)