ComfyUI中MuseTalk插件问题汇总以及解决方法_prompt executed in 16.54 seconds

这里用的是秋叶大佬的ComfyUI的包

问题1：Mcvv的安装中CUDA版本和pytorch版本的不匹配

整合包里面的pytorch的版本对应的CUDA版本是11.8的，如果本机的CUDA版本是12.1的，就得重新装一下pytorch版本，建议使用整合包里面自带的pytorch版本更新（我自己尝试卸载删除会出现后续爆显存的情况）
然后打开右上角的启动命令提示符，输入

where python
1

复制下面这个路径

然后依次在命令行里面安装这几个包(注意换成你自己的路径)

D:\ComfyUI\ComfyUI-aki-v1.3\python\python.exe pip install --no-cache-dir -U openmim 
D:\ComfyUI\ComfyUI-aki-v1.3\python\python.exe mim install mmengine 
D:\ComfyUI\ComfyUI-aki-v1.3\python\python.exe mim install "mmcv>=2.0.1,<2.2.0" 
D:\ComfyUI\ComfyUI-aki-v1.3\python\python.exe mim install "mmdet>=3.1.0" 
D:\ComfyUI\ComfyUI-aki-v1.3\python\python.exe mim install "mmpose>=1.1.0" 
1
2
3
4
5

注意如果遇见下面的情况，等他慢慢下载就是了，不用管他，有时候会很久很久很久很久很久很久很久：命令行里面卡在这一步了Building wheel for mmcv (setup.py)

问题2：没有安装FFmpeg

在ComfyUI里面运行的时候，可能会爆这个错误

2024-05-11 12:01:58,279- root:180- ERROR- Traceback (most recent call last):
  File "D:\ComfyUI\ComfyUI-aki-v1.3\execution.py", line 151, in recursive_execute
    output_data, output_ui = get_output_data(obj, input_data_all)
  File "D:\ComfyUI\ComfyUI-aki-v1.3\execution.py", line 81, in get_output_data
    return_values = map_node_over_list(obj, input_data_all, obj.FUNCTION, allow_interrupt=True)
  File "D:\ComfyUI\ComfyUI-aki-v1.3\execution.py", line 74, in map_node_over_list
    results.append(getattr(obj, func)(**slice_dict(input_data_all, i)))
  File "D:\ComfyUI\ComfyUI-aki-v1.3\custom_nodes\ComfyUI-MuseTalk\nodes.py", line 116, in run
    whisper_feature = audio_processor.audio2feat(audio_path)
  File "D:\ComfyUI\ComfyUI-aki-v1.3/custom_nodes/ComfyUI-MuseTalk\musetalk\whisper\audio2feature.py", line 105, in audio2feat
    result = self.model.transcribe(audio_path)
  File "D:\ComfyUI\ComfyUI-aki-v1.3/custom_nodes/ComfyUI-MuseTalk\musetalk\whisper\whisper\transcribe.py", line 85, in transcribe
    mel = log_mel_spectrogram(audio)
  File "D:\ComfyUI\ComfyUI-aki-v1.3/custom_nodes/ComfyUI-MuseTalk\musetalk\whisper\whisper\audio.py", line 111, in log_mel_spectrogram
    audio = load_audio(audio)
  File "D:\ComfyUI\ComfyUI-aki-v1.3/custom_nodes/ComfyUI-MuseTalk\musetalk\whisper\whisper\audio.py", line 42, in load_audio
    ffmpeg.input(file, threads=0)
  File "D:\ComfyUI\ComfyUI-aki-v1.3\python\lib\site-packages\ffmpeg\_run.py", line 313, in run
    process = run_async(
  File "D:\ComfyUI\ComfyUI-aki-v1.3\python\lib\site-packages\ffmpeg\_run.py", line 284, in run_async
    return subprocess.Popen(
  File "<enhanced_experience vendors.sentry_sdk.integrations.stdlib>", line 217, in sentry_patched_popen_init
  File "D:\ComfyUI\ComfyUI-aki-v1.3\python\lib\subprocess.py", line 971, in __init__
    self._execute_child(args, executable, preexec_fn, close_fds,
  File "D:\ComfyUI\ComfyUI-aki-v1.3\python\lib\subprocess.py", line 1456, in _execute_child
    hp, ht, pid, tid = _winapi.CreateProcess(executable, args,
FileNotFoundError: [WinError 2] 系统找不到指定的文件。

2024-05-11 12:01:58,279- root:124- INFO- Prompt executed in 1.17 seconds
2024-05-11 12:02:12,149- root:451- INFO- got prompt
'ffmpeg' �����ڲ����ⲿ���Ҳ���ǿ����еij���
���������ļ���
2024-05-11 12:02:12,233- root:179- ERROR- !!! Exception during processing!!! [WinError 2] 系统找不到指定的文件。
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34

解决方法：在界面里面直接安装就可以了

问题3：Musetalk的模型缺失

去官网网站里面去下载：
MuseTalk官网
下载后放到这个文件夹里面
ComfyUI/models/diffusers/TMElyralab/MuseTalk/

希望能帮助到大家