[BUG REPORT] ImportError: numpy.core.multiarray failed to import about audio-webui

[gitmylo]

i had updated it twice today, one update made it install a too new numpy version, but it's been changed to use the specific version that i use.

[gitmylo]

It seems you have an outdated numpy version from that error.

The error at the end is just suggesting a common fix, as visual C++ build tools are usually missing, i'll see if i can personalize the message more depending on the error that occurred.

back to the error,

1. run activate_only.bat if you're on windows activate_only.sh if you're on linux or macos.
2. run pip freeze
3. look for numpy
4. it should say something like numpy== and then some numbers. which numbers does it say?

[gitmylo]

Well, i added a numpy minimum version requirement. If you update and run again, it should not crash anymore.

[natlamir]

I had the same error yesterday, I am trying again today, and it was now giving me a different error. This is the error when running run.bat:

Installed audiocraft!
Done installing/checking installs.
Preparing
C:\ai\audio-webui\venv\lib\site-packages\torchaudio\compliance\kaldi.py:22: UserWarning: Failed to initialize NumPy: module compiled against API version 0x10 but this version of numpy is 0xf . Check the section C-API incompatibility at the Troubleshooting ImportError section at https://numpy.org/devdocs/user/troubleshooting-importerror.html#c-api-incompatibility for indications on how to solve this problem . (Triggered internally at ..\torch\csrc\utils\tensor_numpy.cpp:84.)
EPSILON = torch.tensor(torch.finfo(torch.float).eps)
2023-06-29 10:55:50 | INFO | fairseq.tasks.text_to_speech | Please install tensorboardX: pip install tensorboardX
2023-06-29 10:55:50 | WARNING | xformers | A matching Triton is not available, some optimizations will not be enabled.
Error caught was: No module named 'triton'
2023-06-29 10:55:50 | WARNING | xformers | Triton is not available, some optimizations will not be enabled.
This is just a warning: No module named 'triton'
Install finished, pywintypes missing, relaunching script.
Checking installs and venv + autodebug checks
Python version: 3.10.11 (tags/v3.10.11:7d4cc5a, Apr 5 2023, 00:38:17) [MSC v.1929 64 bit (AMD64)]
Installed numpy!

▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄▄
█░▄▄▀██░██░█▄▄░▄▄██░▄▄▄░██░▄▄▀██░▄▄▄██░▄▄▀██░██░██░▄▄░██
█░▀▀░██░██░███░████░███░██░██░██░▄▄▄██░▄▄▀██░██░██░█▀▀██
█░██░██▄▀▀▄███░████░▀▀▀░██░▀▀░██░▀▀▀██░▀▀░██▄▀▀▄██░▀▀▄██
▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀▀

Install failed!
STDOUT:
Requirement already satisfied: numpy in c:\ai\audio-webui\venv\lib\site-packages (1.22.0)
Collecting numpy
Using cached numpy-1.25.0-cp310-cp310-win_amd64.whl (15.0 MB)
Installing collected packages: numpy
Attempting uninstall: numpy
Found existing installation: numpy 1.22.0
Uninstalling numpy-1.22.0:
Successfully uninstalled numpy-1.22.0

STDERR:
ERROR: Could not install packages due to an OSError: [WinError 5] Access is denied: 'C:\ai\audio-webui\venv\Lib\site-packages\~umpy\.libs\libopenblas.EL2C6PLE4ZYW3ECEVIV3OXXGRN2NRFM2.gfortran-win_amd64.dll'
Check the permissions.

[notice] A new release of pip is available: 23.0.1 -> 23.1.2
[notice] To update, run: python.exe -m pip install --upgrade pip

Please read the error above carefully.
If you are unsure, please create an issue at https://github.com/gitmylo/audio-webui/issues.
When creating an issue, please include your full autodebug message.
Exit code: 1

---

Should the script be run as admin? I tried giving Authenticated Users the Full Control permissions for that dll and trying run.bat again, and it gave this error:

Preparing
Traceback (most recent call last):
File "C:\ai\audio-webui\main.py", line 18, in
from webui.modules.implementations.tts_monkeypatching import patch as patch1
File "C:\ai\audio-webui\webui\modules\implementations_init_.py", line 1, in
import webui.modules.implementations.ttsmodels as tts
File "C:\ai\audio-webui\webui\modules\implementations\ttsmodels.py", line 11, in
from TTS.api import TTS
File "C:\ai\audio-webui\venv\lib\site-packages\TTS\api.py", line 13, in
from TTS.utils.audio.numpy_transforms import save_wav
File "C:\ai\audio-webui\venv\lib\site-packages\TTS\utils\audio_init_.py", line 1, in
from TTS.utils.audio.processor import AudioProcessor
File "C:\ai\audio-webui\venv\lib\site-packages\TTS\utils\audio\processor.py", line 10, in
from TTS.utils.audio.numpy_transforms import compute_f0
File "C:\ai\audio-webui\venv\lib\site-packages\TTS\utils\audio\numpy_transforms.py", line 7, in
from librosa import magphase, pyin
File "C:\ai\audio-webui\venv\lib\site-packages\lazy_loader_init_.py", line 77, in getattr
attr = getattr(submod, name)
File "C:\ai\audio-webui\venv\lib\site-packages\lazy_loader_init_.py", line 76, in getattr
submod = importlib.import_module(submod_path)
File "C:\Program Files\Python310\lib\importlib_init_.py", line 126, in import_module
return _bootstrap.gcd_import(name[level:], package, level)
File "C:\ai\audio-webui\venv\lib\site-packages\librosa\core\spectrum.py", line 12, in
from numba import jit
File "C:\ai\audio-webui\venv\lib\site-packages\numba_init.py", line 55, in
ensure_critical_deps()
File "C:\ai\audio-webui\venv\lib\site-packages\numba_init.py", line 42, in _ensure_critical_deps
raise ImportError("Numba needs NumPy 1.24 or less")
ImportError: Numba needs NumPy 1.24 or less
Numba needs NumPy 1.24 or less
Your install might have failed to install one of the requirements, are you missing a package?
You can read common issues at https://github.com/gitmylo/audio-webui/wiki/common-issues

---

I ran activate_only.bat, pip freeze. It says numpy==1.25.0
I ran pip install --upgrade numpy==1.23.5, which gives this error:

(venv) C:\ai\audio-webui>pip install --upgrade numpy==1.23.5
Collecting numpy==1.23.5
Using cached numpy-1.23.5-cp310-cp310-win_amd64.whl (14.6 MB)
Installing collected packages: numpy
Attempting uninstall: numpy
Found existing installation: numpy 1.25.0
Uninstalling numpy-1.25.0:
Successfully uninstalled numpy-1.25.0
ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
tts 0.15.0 requires numpy==1.22.0; python_version <= "3.10", but you have numpy 1.23.5 which is incompatible.
fairseq 0.12.2 requires hydra-core<1.1,>=1.0.7, but you have hydra-core 1.3.2 which is incompatible.
fairseq 0.12.2 requires omegaconf<2.1, but you have omegaconf 2.3.0 which is incompatible.
Successfully installed numpy-1.23.5

[notice] A new release of pip is available: 23.0.1 -> 23.1.2
[notice] To update, run: python.exe -m pip install --upgrade pip

---

However, now when I run run.bat, the application runs successfully after doing the above mentioned things.
Thanks.

[nekogecko2]

STDERR:
ERROR: Could not install packages due to an OSError: [WinError 5] Access is denied: 'C:\ai\audio-webui\venv\Lib\site-packages~umpy.libs\libopenblas.EL2C6PLE4ZYW3ECEVIV3OXXGRN2NRFM2.gfortran-win_amd64.dll'
Check the permissions.

When I reinstalled this program in a new folder, I had all these same things happen to me too.

Running activate.bat -> pip install --upgrade numpy==1.23.5 -> run.bat like you said fixed it

[gitmylo]

STDERR:
ERROR: Could not install packages due to an OSError: [WinError 5] Access is denied: 'C:\ai\audio-webui\venv\Lib\site-packages~umpy.libs\libopenblas.EL2C6PLE4ZYW3ECEVIV3OXXGRN2NRFM2.gfortran-win_amd64.dll'
Check the permissions.

When I reinstalled this program in a new folder, I had all these same things happen to me too.

Running activate.bat -> pip install --upgrade numpy==1.23.5 -> run.bat like you said fixed it

current version doesn't require the --upgrade numpy anymore. as it's included in the auto installer. it will make sure you're on the right version.

[zba]

still have this error with latest code, pip install --upgrade numpy==1.23.5 helped.
You can check this by removing this line from run.sh in my repo https://github.com/LajaSoft/audio-webui-docker and running
docker compose up --build