求助：视频合成报错

2025-03-16 10:23:38 - INFO - App.spawnShell.stdout - platformArch: x86

name: latentsync

######################################################

2025-03-16 10:23:38 - INFO - App.spawnShell.stdout - 2025-03-16 10:23:38 - 开始运行

2025-03-16 10:23:38 - INFO - App.spawnShell.stdout - 2025-03-16 10:23:38 - config {'id': 'VideoGen_2', 'mode': 'local', 'modelConfig': {'type': 'videoGen', 'param': {}, 'video': 'C:\\Users\\Harry\\AppData\\Roaming\\aigcpanel\\data\\videoTemplate\\EJfnqkJ92Uq7Tvg3.mp4', 'audio': 'C:\\Users\\Harry\\AppData\\Roaming\\aigcpanel\\data\\soundClone\\1.wav'}, 'setting': {}} ['D:\\Harry\\AigcPanel\\models\\latentsync\\aigcpanel-server-latentsync-win-x86-v0.1.1\\aigcpanelrun.py', 'C:\\Users\\Harry\\AppData\\Roaming\\aigcpanel\\temp\\file_1742091816759_0xACw8QQhKqFBOKLAwkmrIzkWK8WpC5n.json']

2025-03-16 10:23:43 - INFO - App.spawnShell.stdout - 2025-03-16 10:23:43 - 开始运行 {'UseCuda': False}

Result[VideoGen_2][{"UseCuda": false}]

AigcPanelRunResult[VideoGen_2][eyJVc2VDdWRhIjogZmFsc2V9]

2025-03-16 10:23:57 - INFO - App.spawnShell.stderr - Traceback (most recent call last):

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\runpy.py", line 196, in _run_module_as_main

2025-03-16 10:23:57 - INFO - App.spawnShell.stderr -     return _run_code(code, main_globals, None,

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\runpy.py", line 86, in _run_code

    exec(code, run_globals)

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\scripts\inference.py", line 19, in <module>

    from latentsync.models.unet import UNet3DConditionModel

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\latentsync\models\unet.py", line 27, in <module>

    from ..utils.util import zero_rank_log

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\latentsync\utils\util.py", line 25, in <module>

2025-03-16 10:23:57 - INFO - App.spawnShell.stderr -     import torchvision

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\torchvision\__init__.py", line 10, in <module>

    from torchvision import _meta_registrations, datasets, io, models, ops, transforms, utils  # usort:skip

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\torchvision\models\__init__.py", line 2, in <module>

    from .convnext import *

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\torchvision\models\convnext.py", line 8, in <module>

    from ..ops.misc import Conv2dNormActivation, Permute

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\torchvision\ops\__init__.py", line 23, in <module>

    from .poolers import MultiScaleRoIAlign

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\torchvision\ops\poolers.py", line 10, in <module>

    from .roi_align import roi_align

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\torchvision\ops\roi_align.py", line 7, in <module>

    from torch._dynamo.utils import is_compile_supported

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\torch\_dynamo\__init__.py", line 3, in <module>

2025-03-16 10:23:57 - INFO - App.spawnShell.stderr -     from . import convert_frame, eval_frame, resume_execution

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\torch\_dynamo\convert_frame.py", line 31, in <module>

    from torch._dynamo.utils import CompileTimeInstructionCounter

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\torch\_dynamo\utils.py", line 1320, in <module>

    if has_triton_package():

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\torch\utils\_triton.py", line 9, in has_triton_package

    from triton.compiler.compiler import triton_key

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\triton\__init__.py", line 8, in <module>

    msvc_winsdk_inc_dirs, _ = find_msvc_winsdk()

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\triton\windows_utils.py", line 231, in find_msvc_winsdk

    winsdk_inc_dirs, winsdk_lib_dirs = find_winsdk()

  File "D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\lib\site-packages\triton\windows_utils.py", line 208, in find_winsdk

    winsdk_base_path, version = find_winsdk_registry()

TypeError: cannot unpack non-iterable NoneType object

2025-03-16 10:23:57 - INFO - App.spawnShell.stdout - WARNING: Failed to find MSVC.

WARNING: Failed to find MSVC.

2025-03-16 10:23:58 - INFO - App.spawnShell.stdout - Error: run D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_aienv\python.exe -m scripts.inference --unet_config_path configs/unet/second_stage.yaml --inference_ckpt_path aigcpanelmodels/checkpoints/latentsync_unet.pt --inference_steps 20 --guidance_scale 1.5 --video_path C:\Users\Harry\AppData\Roaming\aigcpanel\data\videoTemplate\EJfnqkJ92Uq7Tvg3.mp4 --audio_path C:\Users\Harry\AppData\Roaming\aigcpanel\data\soundClone\1.wav --video_out_path D:\Harry\AigcPanel\models\latentsync\aigcpanel-server-latentsync-win-x86-v0.1.1\_cache/file\79256f44d34a055a9135e0faa64be53d.mp4 failed: CompletedProcess(args=['D:\\Harry\\AigcPanel\\models\\latentsync\\aigcpanel-server-latentsync-win-x86-v0.1.1\\_aienv\\python.exe', '-m', 'scripts.inference', '--unet_config_path', 'configs/unet/second_stage.yaml', '--inference_ckpt_path', 'aigcpanelmodels/checkpoints/latentsync_unet.pt', '--inference_steps', '20', '--guidance_scale', '1.5', '--video_path', 'C:\\Users\\Harry\\AppData\\Roaming\\aigcpanel\\data\\videoTemplate\\EJfnqkJ92Uq7Tvg3.mp4', '--audio_path', 'C:\\Users\\Harry\\AppData\\Roaming\\aigcpanel\\data\\soundClone\\1.wav', '--video_out_path', 'D:\\Harry\\AigcPanel\\models\\latentsync\\aigcpanel-server-latentsync-win-x86-v0.1.1\\_cache/file\\79256f44d34a055a9135e0faa64be53d.mp4'], returncode=1)

2025-03-16 10:23:59 - INFO - App.spawnShell.stdout - time="2025-03-16T10:23:59+08:00" level=error msg="Error:exit status 0xffffffff"

2025-03-16 10:23:59 - INFO - App.spawnShell.exit - 0

2025-03-16 10:23:59 - ERROR - mapi.server.callFunction.error - {"type":"string","error":"执行失败，请查看模型日志"}

2025-03-16 10:23:59 - ERROR - Task.RunFunc.Error - 执行失败，请查看模型日志

运行过程分析

1. 启动阶段：程序于 2025 年 3 月 16 日 08:44:00 开始启动，接着启动了一个名为 cosyvoice2 - 0.5 的模型服务器，之后该服务器停止运行。

2. 任务尝试：在 10:17:53 和 10:23:36 分别尝试启动 VideoGen_1 和 VideoGen_2 两个视频生成任务。每个任务启动时都会输出服务器启动信息以及任务配置信息，随后便开始运行。

3. 错误出现：在运行这两个任务时，均出现了相同的错误。错误信息显示，程序在导入 torchvision 库时，需要调用 triton 库，而 triton 库在查找 Windows SDK（软件开发工具包）相关信息时出错。具体而言，find_winsdk_registry() 函数返回了 None，这使得程序无法将其解包为预期的两个值，从而引发了 TypeError 错误。

4. 运行失败：由于上述错误，两个视频生成任务均失败，程序输出了错误信息 “执行失败，请查看模型日志”。

求助各位如何解决