search

CarlGao4 / Demucs-Gui

A GUI for music separation AI demucs
GNU General Public License v3.0
566 stars 41 forks source link

Bug: NaN or inf found in stem #90

Closed Sinnaminty closed 2 weeks ago

Sinnaminty commented 2 weeks ago

happen

  1. Attempt to separate a song using the default configuration of htdemucs on 1.3a1_cuda_mkl
  2. Separation completes, but all the stems are completely silent

What is the platform you are using?

Windows

log

2024-10-23 00:25:20,336 (GuiMain.py) (Line 2276) [INFO] : Python version: 3.11.10 | packaged by conda-forge | (main, Sep 30 2024, 17:50:51) [MSC v.1941 64 bit (AMD64)]
2024-10-23 00:25:20,336 (GuiMain.py) (Line 2277) [INFO] : Demucs GUI version: 1.3a1
2024-10-23 00:25:20,408 (GuiMain.py) (Line 2278) [INFO] : System: Windows-10-10.0.22631-SP0
2024-10-23 00:25:20,408 (GuiMain.py) (Line 2279) [INFO] : Architecture: 64bit
2024-10-23 00:25:20,408 (GuiMain.py) (Line 2280) [INFO] : CPU: AMD64 Family 25 Model 33 Stepping 2, AuthenticAMD
2024-10-23 00:25:20,408 (GuiMain.py) (Line 2281) [INFO] : CPU count: 24
2024-10-23 00:25:20,417 (GuiMain.py) (Line 2282) [INFO] : System memory: 51459039232 (47.925GB)
2024-10-23 00:25:20,425 (GuiMain.py) (Line 2285) [INFO] : System free memory: 39733587968 (37.006GB)
2024-10-23 00:25:20,689 (GuiMain.py) (Line 2289) [INFO] : System swap memory: 3221225472 (3GB)
2024-10-23 00:25:20,955 (find_device_win.py) (Line 344) [INFO] : Found GPU: NVIDIA GeForce RTX 4070 SUPER (10DE:2783) Driver version 32.0.15.6590
2024-10-23 00:25:20,955 (GuiMain.py) (Line 2307) [INFO] : Using PySide6
2024-10-23 00:25:20,955 (GuiMain.py) (Line 2308) [INFO] : Qt version: 6.7.3
2024-10-23 00:25:20,955 (GuiMain.py) (Line 2309) [INFO] : PySide6 version: 6.7.3
2024-10-23 00:25:21,015 (shared.py) (Line 491) [INFO] : [1] Thread starter (separator.py) starts
2024-10-23 00:25:21,251 (GuiMain.py) (Line 2314) [DEBUG] : Supported styles: windows11, windowsvista, Windows, Fusion
2024-10-23 00:25:21,253 (GuiMain.py) (Line 2320) [DEBUG] : Current style: windowsvista
intel_extension_for_pytorch\xpu\lazy_init.py:80: UserWarning: XPU Device count is zero! (Triggered internally at E:/AI/ipex-build/2.1.40/intel-extension-for-pytorch/csrc/gpu/runtime/Device.cpp:127.)
  _C._initExtension()
2024-10-23 00:25:24,580 (separator.py) (Line 65) [INFO] : Intel Extension for PyTorch version: 2.1.40+xpu
2024-10-23 00:25:24,580 (separator.py) (Line 65) [INFO] : Intel Extension for PyTorch version: 2.1.40+xpu
2024-10-23 00:25:24,580 (separator.py) (Line 65) [INFO] : Intel Extension for PyTorch version: 2.1.40+xpu
2024-10-23 00:25:24,580 (separator.py) (Line 65) [INFO] : Intel Extension for PyTorch version: 2.1.40+xpu
2024-10-23 00:25:24,580 (separator.py) (Line 65) [INFO] : Intel Extension for PyTorch version: 2.1.40+xpu
2024-10-23 00:25:24,590 (utils.py) (Line 93) [DEBUG] : Attempting to load FFmpeg version 6.
2024-10-23 00:25:24,887 (utils.py) (Line 130) [DEBUG] : Failed to load FFmpeg 6 extension.
Traceback (most recent call last):
  File "torchaudio\_extension\utils.py", line 128, in _find_ffmpeg_extension
    return _find_versionsed_ffmpeg_extension(ffmpeg_ver)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "torchaudio\_extension\utils.py", line 113, in _find_versionsed_ffmpeg_extension
    _try_access_avutil(ffmpeg_ver)
  File "torchaudio\_extension\utils.py", line 89, in _try_access_avutil
    torchaudio.lib._torchaudio.find_avutil(libavutil)
RuntimeError: error in LoadLibrary for avutil-58.dll. WinError 126: The specified module could not be found.

2024-10-23 00:25:24,888 (utils.py) (Line 93) [DEBUG] : Attempting to load FFmpeg version 5.
2024-10-23 00:25:24,890 (utils.py) (Line 130) [DEBUG] : Failed to load FFmpeg 5 extension.
Traceback (most recent call last):
  File "torchaudio\_extension\utils.py", line 128, in _find_ffmpeg_extension
    return _find_versionsed_ffmpeg_extension(ffmpeg_ver)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "torchaudio\_extension\utils.py", line 113, in _find_versionsed_ffmpeg_extension
    _try_access_avutil(ffmpeg_ver)
  File "torchaudio\_extension\utils.py", line 89, in _try_access_avutil
    torchaudio.lib._torchaudio.find_avutil(libavutil)
RuntimeError: error in LoadLibrary for avutil-57.dll. WinError 126: The specified module could not be found.

2024-10-23 00:25:24,891 (utils.py) (Line 93) [DEBUG] : Attempting to load FFmpeg version 4.
2024-10-23 00:25:24,893 (utils.py) (Line 130) [DEBUG] : Failed to load FFmpeg 4 extension.
Traceback (most recent call last):
  File "torchaudio\_extension\utils.py", line 128, in _find_ffmpeg_extension
    return _find_versionsed_ffmpeg_extension(ffmpeg_ver)
           ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "torchaudio\_extension\utils.py", line 113, in _find_versionsed_ffmpeg_extension
    _try_access_avutil(ffmpeg_ver)
  File "torchaudio\_extension\utils.py", line 89, in _try_access_avutil
    torchaudio.lib._torchaudio.find_avutil(libavutil)
RuntimeError: error in LoadLibrary for avutil-56.dll. WinError 126: The specified module could not be found.

2024-10-23 00:25:24,893 (__init__.py) (Line 96) [DEBUG] : Failed to initialize ffmpeg bindings
Traceback (most recent call last):
  File "torchaudio\_extension\__init__.py", line 89, in <module>
    _FFMPEG_EXT = _init_ffmpeg()
                  ^^^^^^^^^^^^^^
  File "torchaudio\_extension\utils.py", line 162, in _init_ffmpeg
    ext = _find_ffmpeg_extension(ffmpeg_vers, show_error)
          ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "torchaudio\_extension\utils.py", line 132, in _find_ffmpeg_extension
    raise ImportError(f"Failed to intialize FFmpeg extension. Tried versions: {ffmpeg_vers}")
ImportError: Failed to intialize FFmpeg extension. Tried versions: ['6', '5', '4']
PyInstaller\loader\pyimod02_importers.py:378: UserWarning: TorchAudio's global backend is now deprecated. Please enable distpatcher by setting `TORCHAUDIO_USE_BACKEND_DISPATCHER=1`, and specify backend when calling load/info/save function.
2024-10-23 00:25:25,057 (audio.py) (Line 36) [INFO] : Soundfile version: 0.12.1
2024-10-23 00:25:25,057 (audio.py) (Line 37) [INFO] : libsndfile version: 1.2.0
2024-10-23 00:25:25,057 (audio.py) (Line 38) [INFO] : SoXR version: 0.5.0.post1
2024-10-23 00:25:25,057 (audio.py) (Line 39) [INFO] : libsoxr version: 0.1.3-11-gedbdb40
2024-10-23 00:25:25,058 (separator.py) (Line 89) [INFO] : Demucs version: 4.1.0a3
2024-10-23 00:25:25,058 (separator.py) (Line 90) [INFO] : PyTorch version: 2.1.0a0+cu118
2024-10-23 00:25:25,058 (separator.py) (Line 100) [INFO] : CUDA Info: _CudaDeviceProperties(name='NVIDIA GeForce RTX 4070 SUPER', major=8, minor=9, total_memory=12281MB, multi_processor_count=56)
2024-10-23 00:25:25,058 (separator.py) (Line 104) [INFO] : CUDA Arch list: ['sm_35', 'sm_50', 'sm_60', 'sm_61', 'sm_70', 'sm_75', 'sm_80', 'sm_86', 'sm_89', 'sm_90']
2024-10-23 00:25:26,091 (audio.py) (Line 54) [INFO] : ffmpeg -version output:
ffmpeg version 6.1.1-full_build-www.gyan.dev Copyright (c) 2000-2023 the FFmpeg developers
built with gcc 12.2.0 (Rev10, Built by MSYS2 project)
configuration: --enable-gpl --enable-version3 --enable-shared --disable-w32threads --disable-autodetect --enable-fontconfig --enable-iconv --enable-gnutls --enable-libxml2 --enable-gmp --enable-bzlib --enable-lzma --enable-libsnappy --enable-zlib --enable-librist --enable-libsrt --enable-libssh --enable-libzmq --enable-avisynth --enable-libbluray --enable-libcaca --enable-sdl2 --enable-libaribb24 --enable-libaribcaption --enable-libdav1d --enable-libdavs2 --enable-libuavs3d --enable-libzvbi --enable-librav1e --enable-libsvtav1 --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxavs2 --enable-libxvid --enable-libaom --enable-libjxl --enable-libopenjpeg --enable-libvpx --enable-mediafoundation --enable-libass --enable-frei0r --enable-libfreetype --enable-libfribidi --enable-libharfbuzz --enable-liblensfun --enable-libvidstab --enable-libvmaf --enable-libzimg --enable-amf --enable-cuda-llvm --enable-cuvid --enable-ffnvcodec --enable-nvdec --enable-nvenc --enable-dxva2 --enable-d3d11va --enable-libvpl --enable-libshaderc --enable-vulkan --enable-libplacebo --enable-opencl --enable-libcdio --enable-libgme --enable-libmodplug --enable-libopenmpt --enable-libopencore-amrwb --enable-libmp3lame --enable-libshine --enable-libtheora --enable-libtwolame --enable-libvo-amrwbenc --enable-libcodec2 --enable-libilbc --enable-libgsm --enable-libopencore-amrnb --enable-libopus --enable-libspeex --enable-libvorbis --enable-ladspa --enable-libbs2b --enable-libflite --enable-libmysofa --enable-librubberband --enable-libsoxr --enable-chromaprint
libavutil      58. 29.100 / 58. 29.100
libavcodec     60. 31.102 / 60. 31.102
libavformat    60. 16.100 / 60. 16.100
libavdevice    60.  3.100 / 60.  3.100
libavfilter     9. 12.100 /  9. 12.100
libswscale      7.  5.100 /  7.  5.100
libswresample   4. 12.100 /  4. 12.100
libpostproc    57.  3.100 / 57.  3.100

2024-10-23 00:25:26,091 (audio.py) (Line 57) [INFO] : SoXR enabled in FFmpeg
2024-10-23 00:25:26,124 (audio.py) (Line 62) [INFO] : Using ffmpeg from C:\Users\aiden\Downloads\Demucs-GUI_1.3a1_cuda_mkl\Demucs-GUI_1.3a1_cuda_mkl\ffmpeg\ffmpeg.EXE
2024-10-23 00:25:26,124 (audio.py) (Line 63) [INFO] : ffprobe -version output:
ffprobe version 6.1.1-full_build-www.gyan.dev Copyright (c) 2007-2023 the FFmpeg developers
built with gcc 12.2.0 (Rev10, Built by MSYS2 project)
configuration: --enable-gpl --enable-version3 --enable-shared --disable-w32threads --disable-autodetect --enable-fontconfig --enable-iconv --enable-gnutls --enable-libxml2 --enable-gmp --enable-bzlib --enable-lzma --enable-libsnappy --enable-zlib --enable-librist --enable-libsrt --enable-libssh --enable-libzmq --enable-avisynth --enable-libbluray --enable-libcaca --enable-sdl2 --enable-libaribb24 --enable-libaribcaption --enable-libdav1d --enable-libdavs2 --enable-libuavs3d --enable-libzvbi --enable-librav1e --enable-libsvtav1 --enable-libwebp --enable-libx264 --enable-libx265 --enable-libxavs2 --enable-libxvid --enable-libaom --enable-libjxl --enable-libopenjpeg --enable-libvpx --enable-mediafoundation --enable-libass --enable-frei0r --enable-libfreetype --enable-libfribidi --enable-libharfbuzz --enable-liblensfun --enable-libvidstab --enable-libvmaf --enable-libzimg --enable-amf --enable-cuda-llvm --enable-cuvid --enable-ffnvcodec --enable-nvdec --enable-nvenc --enable-dxva2 --enable-d3d11va --enable-libvpl --enable-libshaderc --enable-vulkan --enable-libplacebo --enable-opencl --enable-libcdio --enable-libgme --enable-libmodplug --enable-libopenmpt --enable-libopencore-amrwb --enable-libmp3lame --enable-libshine --enable-libtheora --enable-libtwolame --enable-libvo-amrwbenc --enable-libcodec2 --enable-libilbc --enable-libgsm --enable-libopencore-amrnb --enable-libopus --enable-libspeex --enable-libvorbis --enable-ladspa --enable-libbs2b --enable-libflite --enable-libmysofa --enable-librubberband --enable-libsoxr --enable-chromaprint
libavutil      58. 29.100 / 58. 29.100
libavcodec     60. 31.102 / 60. 31.102
libavformat    60. 16.100 / 60. 16.100
libavdevice    60.  3.100 / 60.  3.100
libavfilter     9. 12.100 /  9. 12.100
libswscale      7.  5.100 /  7.  5.100
libswresample   4. 12.100 /  4. 12.100
libpostproc    57.  3.100 / 57.  3.100

2024-10-23 00:25:26,189 (audio.py) (Line 84) [INFO] : FFmpeg protocols: async, bluray, cache, concat, concatf, crypto, data, fd, ffrtmpcrypt, ffrtmphttp, file, ftp, gopher, gophers, hls, http, httpproxy, https, ipfs, ipns, mmsh, mmst, pipe, rist, rtmp, rtmpe, rtmps, rtmpt, rtmpte, rtmpts, rtp, sftp, srt, srtp, subfile, tcp, tls, udp, udplite, zmq
2024-10-23 00:25:27,189 (shared.py) (Line 508) [INFO] : [1] Thread starter (separator.py) ends
2024-10-23 00:25:27,194 (shared.py) (Line 491) [INFO] : [2] Thread checkUpdate (shared.py) starts
2024-10-23 00:25:27,206 (shared.py) (Line 524) [INFO] : Checking for updates...
2024-10-23 00:25:27,904 (shared.py) (Line 527) [INFO] : Latest version: 1.3a1
2024-10-23 00:25:27,905 (shared.py) (Line 508) [INFO] : [2] Thread checkUpdate (shared.py) ends
2024-10-23 00:25:49,956 (shared.py) (Line 491) [INFO] : [3] Thread loadModel (GuiMain.py) starts
2024-10-23 00:25:49,958 (GuiMain.py) (Line 701) [INFO] : Loading model htdemucs from repo C:\Users\aiden\Downloads\Demucs-GUI_1.3a1_cuda_mkl\Demucs-GUI_1.3a1_cuda_mkl\pretrained
2024-10-23 00:25:50,423 (GuiMain.py) (Line 729) [INFO] : Model loaded within 0.4620s
2024-10-23 00:25:50,423 (GuiMain.py) (Line 730) [INFO] : Model: htdemucs
Repo: C:\Users\aiden\Downloads\Demucs-GUI_1.3a1_cuda_mkl\Demucs-GUI_1.3a1_cuda_mkl\pretrained
Type: Bag of models
Audio channels: 2
Sample rate: 44100
Sources: drums, bass, other, vocals

Model 0:
　Type: HTDemucs
　Default segment: 7.8
　Weight: [1.0, 1.0, 1.0, 1.0]
2024-10-23 00:25:50,436 (shared.py) (Line 508) [INFO] : [3] Thread loadModel (GuiMain.py) ends
2024-10-23 00:25:50,442 (shared.py) (Line 296) [DEBUG] : (switchEncoder) Set history ('encoder',) to 0
2024-10-23 00:25:50,442 (shared.py) (Line 298) [DEBUG] : History not changed, ignored
2024-10-23 00:25:50,444 (GuiMain.py) (Line 1737) [INFO] : Preset stem key: ["bass","drums","other","vocals"]
2024-10-23 00:25:50,444 (GuiMain.py) (Line 1892) [INFO] : Adding saved presets: ['meow']
2024-10-23 00:25:50,451 (GuiMain.py) (Line 1989) [INFO] : Applying preset meow
2024-10-23 00:25:50,464 (shared.py) (Line 491) [INFO] : [4] Thread loadURLname_thread (GuiMain.py) starts
2024-10-23 00:26:24,803 (shared.py) (Line 491) [INFO] : [5] Thread separate (separator.py) starts
2024-10-23 00:26:24,803 (separator.py) (Line 461) [INFO] : Start separating audio: Armed & Dangerous.mp3
2024-10-23 00:26:24,803 (separator.py) (Line 462) [INFO] : Parameters: segment=2.80 overlap=0.05 shifts=0
2024-10-23 00:26:24,803 (separator.py) (Line 463) [INFO] : Device: cuda:0
2024-10-23 00:26:24,803 (audio.py) (Line 97) [DEBUG] : Reading audio with soundfile: C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\Armed & Dangerous.mp3
2024-10-23 00:26:25,020 (audio.py) (Line 113) [INFO] : Read audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\Armed & Dangerous.mp3: samplerate=44100 shape=(10345271, 2)
2024-10-23 00:26:25,443 (separator.py) (Line 499) [INFO] : Running separation...
2024-10-23 00:26:35,419 (separator.py) (Line 526) [INFO] : Saving separated audio...
2024-10-23 00:26:35,422 (shared.py) (Line 491) [INFO] : [6] Thread save (GuiMain.py) starts
2024-10-23 00:26:35,425 (shared.py) (Line 508) [INFO] : [5] Thread separate (separator.py) ends
2024-10-23 00:26:35,435 (shared.py) (Line 296) [DEBUG] : (save) Set history ('save_location',) to OrderedSet(['separated/{model}/{track}/{stem}.{ext}', 'separated/{track}/{stem}.{ext}'])
2024-10-23 00:26:35,440 (GuiMain.py) (Line 2037) [INFO] : Mixing stem drums
2024-10-23 00:26:36,688 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem drums
2024-10-23 00:26:37,156 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\drums.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:26:37,156 (GuiMain.py) (Line 2037) [INFO] : Mixing stem bass
2024-10-23 00:26:38,384 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem bass
2024-10-23 00:26:38,881 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\bass.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:26:38,881 (GuiMain.py) (Line 2037) [INFO] : Mixing stem other
2024-10-23 00:26:40,118 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem other
2024-10-23 00:26:40,607 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\other.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:26:40,607 (GuiMain.py) (Line 2037) [INFO] : Mixing stem vocals
2024-10-23 00:26:41,883 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem vocals
2024-10-23 00:26:42,357 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\vocals.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:26:42,357 (GuiMain.py) (Line 2037) [INFO] : Mixing stem master
2024-10-23 00:26:43,610 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem master
2024-10-23 00:26:44,086 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\master.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:26:44,086 (shared.py) (Line 508) [INFO] : [6] Thread save (GuiMain.py) ends
2024-10-23 00:27:39,573 (shared.py) (Line 296) [DEBUG] : (<lambda>) Set history ('clip_mode',) to none
2024-10-23 00:28:22,792 (shared.py) (Line 491) [INFO] : [7] Thread separate (separator.py) starts
2024-10-23 00:28:22,792 (separator.py) (Line 461) [INFO] : Start separating audio: Armed & Dangerous.mp3
2024-10-23 00:28:22,792 (separator.py) (Line 462) [INFO] : Parameters: segment=7.80 overlap=0.25 shifts=0
2024-10-23 00:28:22,792 (separator.py) (Line 463) [INFO] : Device: cuda:0
2024-10-23 00:28:22,792 (audio.py) (Line 97) [DEBUG] : Reading audio with soundfile: C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\Armed & Dangerous.mp3
2024-10-23 00:28:23,001 (audio.py) (Line 113) [INFO] : Read audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\Armed & Dangerous.mp3: samplerate=44100 shape=(10345271, 2)
2024-10-23 00:28:23,446 (separator.py) (Line 499) [INFO] : Running separation...
2024-10-23 00:28:29,575 (separator.py) (Line 526) [INFO] : Saving separated audio...
2024-10-23 00:28:29,578 (shared.py) (Line 491) [INFO] : [8] Thread save (GuiMain.py) starts
2024-10-23 00:28:29,581 (shared.py) (Line 508) [INFO] : [7] Thread separate (separator.py) ends
2024-10-23 00:28:29,591 (shared.py) (Line 296) [DEBUG] : (save) Set history ('save_location',) to OrderedSet(['separated/{model}/{track}/{stem}.{ext}', 'separated/{track}/{stem}.{ext}'])
2024-10-23 00:28:29,598 (GuiMain.py) (Line 2037) [INFO] : Mixing stem drums
2024-10-23 00:28:30,902 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem drums
2024-10-23 00:28:31,002 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\drums.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:28:31,002 (GuiMain.py) (Line 2037) [INFO] : Mixing stem bass
2024-10-23 00:28:32,418 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem bass
2024-10-23 00:28:32,531 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\bass.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:28:32,531 (GuiMain.py) (Line 2037) [INFO] : Mixing stem other
2024-10-23 00:28:33,769 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem other
2024-10-23 00:28:33,872 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\other.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:28:33,872 (GuiMain.py) (Line 2037) [INFO] : Mixing stem vocals
2024-10-23 00:28:35,118 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem vocals
2024-10-23 00:28:35,219 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\vocals.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:28:35,219 (GuiMain.py) (Line 2037) [INFO] : Mixing stem master
2024-10-23 00:28:36,513 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem master
2024-10-23 00:28:36,612 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\master.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:28:36,613 (shared.py) (Line 508) [INFO] : [8] Thread save (GuiMain.py) ends
2024-10-23 00:29:20,033 (shared.py) (Line 296) [DEBUG] : (<lambda>) Set history ('clip_mode',) to rescale
2024-10-23 00:29:23,949 (shared.py) (Line 296) [DEBUG] : (switchEncoder) Set history ('encoder',) to 1
2024-10-23 00:29:48,438 (shared.py) (Line 296) [DEBUG] : (switchEncoder) Set history ('encoder',) to 0
2024-10-23 00:29:51,614 (shared.py) (Line 296) [DEBUG] : (<lambda>) Set history ('sample_fmt',) to float32
2024-10-23 00:30:01,336 (shared.py) (Line 491) [INFO] : [9] Thread separate (separator.py) starts
2024-10-23 00:30:01,336 (separator.py) (Line 461) [INFO] : Start separating audio: Armed & Dangerous.mp3
2024-10-23 00:30:01,336 (separator.py) (Line 462) [INFO] : Parameters: segment=7.80 overlap=0.25 shifts=0
2024-10-23 00:30:01,336 (separator.py) (Line 463) [INFO] : Device: cuda:0
2024-10-23 00:30:01,336 (audio.py) (Line 97) [DEBUG] : Reading audio with soundfile: C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\Armed & Dangerous.mp3
2024-10-23 00:30:01,548 (audio.py) (Line 113) [INFO] : Read audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\Armed & Dangerous.mp3: samplerate=44100 shape=(10345271, 2)
2024-10-23 00:30:01,993 (separator.py) (Line 499) [INFO] : Running separation...
2024-10-23 00:30:08,271 (separator.py) (Line 526) [INFO] : Saving separated audio...
2024-10-23 00:30:08,273 (shared.py) (Line 491) [INFO] : [10] Thread save (GuiMain.py) starts
2024-10-23 00:30:08,278 (shared.py) (Line 508) [INFO] : [9] Thread separate (separator.py) ends
2024-10-23 00:30:08,288 (shared.py) (Line 296) [DEBUG] : (save) Set history ('save_location',) to OrderedSet(['separated/{model}/{track}/{stem}.{ext}', 'separated/{track}/{stem}.{ext}'])
2024-10-23 00:30:08,296 (GuiMain.py) (Line 2037) [INFO] : Mixing stem drums
2024-10-23 00:30:09,600 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem drums
2024-10-23 00:30:10,087 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\drums.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:30:10,088 (GuiMain.py) (Line 2037) [INFO] : Mixing stem bass
2024-10-23 00:30:11,323 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem bass
2024-10-23 00:30:11,859 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\bass.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:30:11,859 (GuiMain.py) (Line 2037) [INFO] : Mixing stem other
2024-10-23 00:30:13,064 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem other
2024-10-23 00:30:13,544 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\other.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:30:13,545 (GuiMain.py) (Line 2037) [INFO] : Mixing stem vocals
2024-10-23 00:30:14,814 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem vocals
2024-10-23 00:30:15,292 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\vocals.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:30:15,292 (GuiMain.py) (Line 2037) [INFO] : Mixing stem master
2024-10-23 00:30:16,553 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem master
2024-10-23 00:30:17,022 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\master.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:30:17,022 (shared.py) (Line 508) [INFO] : [10] Thread save (GuiMain.py) ends
2024-10-23 00:31:06,377 (shared.py) (Line 296) [DEBUG] : (<lambda>) Set history ('sample_fmt',) to int24
2024-10-23 00:31:13,617 (shared.py) (Line 491) [INFO] : [11] Thread separate (separator.py) starts
2024-10-23 00:31:13,617 (separator.py) (Line 461) [INFO] : Start separating audio: ATW - 100%.mp3
2024-10-23 00:31:13,617 (separator.py) (Line 462) [INFO] : Parameters: segment=7.80 overlap=0.25 shifts=0
2024-10-23 00:31:13,617 (separator.py) (Line 463) [INFO] : Device: cuda:0
2024-10-23 00:31:13,617 (audio.py) (Line 97) [DEBUG] : Reading audio with soundfile: C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\ATW - 100%.mp3
2024-10-23 00:31:13,945 (audio.py) (Line 113) [INFO] : Read audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\ATW - 100%.mp3: samplerate=44100 shape=(16028901, 2)
2024-10-23 00:31:14,634 (separator.py) (Line 499) [INFO] : Running separation...
2024-10-23 00:31:23,939 (separator.py) (Line 526) [INFO] : Saving separated audio...
2024-10-23 00:31:23,941 (shared.py) (Line 491) [INFO] : [12] Thread save (GuiMain.py) starts
2024-10-23 00:31:23,947 (shared.py) (Line 508) [INFO] : [11] Thread separate (separator.py) ends
2024-10-23 00:31:23,957 (shared.py) (Line 296) [DEBUG] : (save) Set history ('save_location',) to OrderedSet(['separated/{model}/{track}/{stem}.{ext}', 'separated/{track}/{stem}.{ext}'])
2024-10-23 00:31:23,965 (GuiMain.py) (Line 2037) [INFO] : Mixing stem drums
2024-10-23 00:31:26,092 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem drums
2024-10-23 00:31:26,794 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\ATW - 100%\drums.wav: shape=torch.Size([2, 16028901])
2024-10-23 00:31:26,794 (GuiMain.py) (Line 2037) [INFO] : Mixing stem bass
2024-10-23 00:31:28,891 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem bass
2024-10-23 00:31:29,634 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\ATW - 100%\bass.wav: shape=torch.Size([2, 16028901])
2024-10-23 00:31:29,634 (GuiMain.py) (Line 2037) [INFO] : Mixing stem other
2024-10-23 00:31:31,702 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem other
2024-10-23 00:31:32,335 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\ATW - 100%\other.wav: shape=torch.Size([2, 16028901])
2024-10-23 00:31:32,336 (GuiMain.py) (Line 2037) [INFO] : Mixing stem vocals
2024-10-23 00:31:34,411 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem vocals
2024-10-23 00:31:35,113 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\ATW - 100%\vocals.wav: shape=torch.Size([2, 16028901])
2024-10-23 00:31:35,113 (GuiMain.py) (Line 2037) [INFO] : Mixing stem master
2024-10-23 00:31:37,009 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem master
2024-10-23 00:31:37,658 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\ATW - 100%\master.wav: shape=torch.Size([2, 16028901])
2024-10-23 00:31:37,673 (shared.py) (Line 508) [INFO] : [12] Thread save (GuiMain.py) ends
2024-10-23 00:34:49,024 (shared.py) (Line 296) [DEBUG] : (<lambda>) Set history ('sample_fmt',) to int16
2024-10-23 00:35:11,257 (shared.py) (Line 491) [INFO] : [13] Thread separate (separator.py) starts
2024-10-23 00:35:11,258 (separator.py) (Line 461) [INFO] : Start separating audio: Armed & Dangerous.mp3
2024-10-23 00:35:11,258 (separator.py) (Line 462) [INFO] : Parameters: segment=7.80 overlap=0.25 shifts=0
2024-10-23 00:35:11,258 (separator.py) (Line 463) [INFO] : Device: cuda:0
2024-10-23 00:35:11,258 (audio.py) (Line 97) [DEBUG] : Reading audio with soundfile: C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\Armed & Dangerous.mp3
2024-10-23 00:35:11,467 (audio.py) (Line 113) [INFO] : Read audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\Armed & Dangerous.mp3: samplerate=44100 shape=(10345271, 2)
2024-10-23 00:35:11,896 (separator.py) (Line 499) [INFO] : Running separation...
2024-10-23 00:35:18,372 (separator.py) (Line 526) [INFO] : Saving separated audio...
2024-10-23 00:35:18,374 (shared.py) (Line 491) [INFO] : [14] Thread save (GuiMain.py) starts
2024-10-23 00:35:18,379 (shared.py) (Line 508) [INFO] : [13] Thread separate (separator.py) ends
2024-10-23 00:35:18,390 (shared.py) (Line 296) [DEBUG] : (save) Set history ('save_location',) to OrderedSet(['separated/{model}/{track}/{stem}.{ext}', 'separated/{track}/{stem}.{ext}'])
2024-10-23 00:35:18,398 (GuiMain.py) (Line 2037) [INFO] : Mixing stem drums
2024-10-23 00:35:19,791 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem drums
2024-10-23 00:35:20,293 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\drums.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:35:20,293 (GuiMain.py) (Line 2037) [INFO] : Mixing stem bass
2024-10-23 00:35:21,736 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem bass
2024-10-23 00:35:22,242 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\bass.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:35:22,242 (GuiMain.py) (Line 2037) [INFO] : Mixing stem other
2024-10-23 00:35:23,571 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem other
2024-10-23 00:35:24,039 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\other.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:35:24,040 (GuiMain.py) (Line 2037) [INFO] : Mixing stem vocals
2024-10-23 00:35:25,361 (GuiMain.py) (Line 1166) [WARNING] : NaN or inf found in stem vocals
2024-10-23 00:35:25,839 (audio.py) (Line 162) [INFO] : Saved audio C:\Users\aiden\Desktop\ALL_MUSIC_STEMS\UKG\separated\htdemucs\Armed & Dangerous\vocals.wav: shape=torch.Size([2, 10345271])
2024-10-23 00:35:25,839 (shared.py) (Line 508) [INFO] : [14] Thread save (GuiMain.py) ends
2024-10-23 00:36:24,667 (GuiMain.py) (Line 1989) [INFO] : Applying preset meow
CarlGao4 commented 2 weeks ago

Can you provide your sample audio file?

Sinnaminty commented 2 weeks ago

Armed & Dangerous.zip

On my laptop I am able to successfully separate this same audio file using the CPU version of Demucs, so I am willing to bet it is not the audio file

CarlGao4 commented 2 weeks ago

It's due to intel-extension-for-pytorch I don't know why, but sometimes it may cause calculation errors (just like 1 + 1 = 3) I'll upload the binaries again tomorrow