[Bug] UserWarning: Failed to initialize NumPy: module compiled against API version 0xf but this version of numpy is 0xe (Triggered internally at ..\torch\csrc\utils\tensor_numpy.cpp:68.)

Describe the bug

TTS requires “numpy-1.21.6” but this time torch cries with the above error

if I upgrade numpy to 1.22.4, this time torch works but TTS cries:

``pkg_resources.DistributionNotFound: The ‘numpy==1.21.6’ distribution was not found and is required by TTS````

How can I fix this dilemma?

To Reproduce

Typing

tts --list_models

in CMD window

Expected behavior

No response

Logs

No response

Environment

{
    "CUDA": {
        "GPU": [],
        "available": false,
        "version": null
    },
    "Packages": {
        "PyTorch_debug": false,
        "PyTorch_version": "1.11.0+cpu",
        "TTS": "0.6.2",
        "numpy": "1.21.6"
    },
    "System": {
        "OS": "Windows",
        "architecture": [
            "64bit",
            "WindowsPE"
        ],
        "processor": "Intel64 Family 6 Model 165 Stepping 2, GenuineIntel",
        "python": "3.10.4",
        "version": "10.0.22000"
    }
}

Additional context

No response

[wasertech]

Wait, no numba on py310 has numpy requirement of: Requirement: numpy<1.23,>=1.18 in /home/waser/Projets/Données/TTS/venv-tts/lib/python3.10/site-packages (from numba) (1.21.6) So 1.21.6 should work perfectly, yet here we are…

[wasertech]

It may not be a bug but it’s an issue. We should keep track of it.

❯ tts --text "This is a wonderful day, and I think I'm going to celebrate with a cookie" --model_name tts_models/en/vctk/vits --speaker_idx p335
RuntimeError: module compiled against API version 0xf but this version of numpy is 0xe
Traceback (most recent call last):
  File "/home/waser/.local/bin/tts", line 5, in <module>
    from TTS.bin.synthesize import main
  File "/home/waser/.local/lib/python3.10/site-packages/TTS/bin/synthesize.py", line 12, in <module>
    from TTS.utils.synthesizer import Synthesizer
  File "/home/waser/.local/lib/python3.10/site-packages/TTS/utils/synthesizer.py", line 14, in <module>
    from TTS.utils.audio import AudioProcessor
  File "/home/waser/.local/lib/python3.10/site-packages/TTS/utils/audio.py", line 3, in <module>
    import librosa
  File "/home/waser/.local/lib/python3.10/site-packages/librosa/__init__.py", line 211, in <module>
    from . import core
  File "/home/waser/.local/lib/python3.10/site-packages/librosa/core/__init__.py", line 5, in <module>
    from .convert import *  # pylint: disable=wildcard-import
  File "/home/waser/.local/lib/python3.10/site-packages/librosa/core/convert.py", line 7, in <module>
    from . import notation
  File "/home/waser/.local/lib/python3.10/site-packages/librosa/core/notation.py", line 8, in <module>
    from ..util.exceptions import ParameterError
  File "/home/waser/.local/lib/python3.10/site-packages/librosa/util/__init__.py", line 83, in <module>
    from .utils import *  # pylint: disable=wildcard-import
  File "/home/waser/.local/lib/python3.10/site-packages/librosa/util/utils.py", line 6, in <module>
    import scipy.ndimage
  File "/usr/lib/python3.10/site-packages/scipy/ndimage/__init__.py", line 151, in <module>
    from ._filters import *  # noqa: F401 F403
  File "/usr/lib/python3.10/site-packages/scipy/ndimage/_filters.py", line 37, in <module>
    from . import _nd_image
ImportError: numpy.core.multiarray failed to import

Using venv with python 3.10.5.

TTS v0.7.1 needs (according to requirements.txt):

• numba==0.55.1
• numpy==1.21.6

Numba needs numpy >=1.11, <=1.18 (according to it’s setup.py)

See the issue? [Numba]numpy==1.18 < [TTS]numpy==1.21.6