wudicgi / spleetermsvcexe Goto Github PK

Will you add GPU support in the future?
I currently have an i7 870 and a GTX 1050 Ti.

When I try to process an audio with SpleeterMsvcExe, I get this error:
Error: The current processor lacks AVX support, the program will exit.
I guess it currently only uses the CPU.

SpleeterGui
It does start the process, however it increases the CPU temperature considerably, so I prefer to avoid it.
By the way, maybe it would be better to change the name of the interface for SpleeterMsvcExe to something more differentiating?
Like Spleeter UI, Win Spleeter, Wudi Spleeter or anything else instead of “Spleeter GUI” :p

StemRoller
works without any problems thanks to CUDA support, without affecting the CPU.
This one is based on Demucs.

Demucs-Gui
Oddly enough this one for some reason gives results even faster than StemRoller, although its interface is much less user friendly.

https://wwi.lanzoup.com/i81Iq0edhxzi

你好，博主，感谢你提供这样的工具。不过，我在使用时报错了，能看下吗？

E:\AI\SpleeterMsvcExe-v1
λ Spleeter.exe test.mp3
Input file:
test.mp3

Output files:
test.vocals.mp3
test.accompaniment.mp3

[  0.00%] Reading audio samples, 0/3359232
[  1.00%] Reading audio samples, 1679663/3359232
[  2.00%] Reading audio samples, 3358127/3358127
[  2.00%] Loading spleeter model, 0/1
2023-05-23 14:53:19.715000: I tensorflow/cc/saved_model/reader.cc:31] Reading SavedModel from: E:\AI\SpleeterMsvcExe-v1\models\2stems
2023-05-23 14:53:19.740000: I tensorflow/cc/saved_model/reader.cc:54] Reading meta graph with tags { serve }
2023-05-23 14:53:19.763000: I tensorflow/core/platform/cpu_feature_guard.cc:142] Your CPU supports instructions that this TensorFlow binary was not compiled to use: AVX2
2023-05-23 14:53:19.838000: I tensorflow/cc/saved_model/loader.cc:202] Restoring SavedModel bundle.
2023-05-23 14:53:20.184000: I tensorflow/cc/saved_model/loader.cc:151] Running initialization op on SavedModel bundle at path: E:\AI\SpleeterMsvcExe-v1\models\2stems
2023-05-23 14:53:20.267000: I tensorflow/cc/saved_model/loader.cc:311] SavedModel load for tags { serve }; Status: success. Took 551000 microseconds.
[  5.00%] Loading spleeter model, 1/1
[ 38.49%] Processing segment, 1323000/3358127
[ 71.97%] Processing segment, 2646000/3358127
[ 90.00%] Processing segment, 3358127/3358127
[mp3_mf @ 000000000056eb80] could not find any MFT for the given media type
[mp3_mf @ 000000000056eb80] could not create MFT
Error: Failed to write output file "test.vocals.mp3".

E:\AI\SpleeterMsvcExe-v1
λ

我是在cmder命令行下运行的。
请问博主，该如何解决此错误？

Hi, thanks for great effort about SpleeterMsvcExe :-) It's easy and fast :-) Can you add the Demucs 3 algorithm to your project?
The Demucs 3 has high quality separation model B. Please check here

An example project is written in python : here
and main website here

I hope, you'll do it,
Best regards,

No effect at all, no output to the file
win11 22621 x64

How do I Export Spleeter Models for SpleeterMsvcExe to use?
I tried to use the Python code of Export Spleeter Models mentioned in Guillaume Vincke's spleeterpp project to Export Spleeter Models for SpleeterMsvcExe to use, but SpleeterMsvcExe could not accept it! And No Error appear!
Because I need F=2048 Spleeter Models to test, Thank you!