install seems broken #1306

remic33 · 2023-03-30T13:06:15Z

Hi everyone,
I am trying to use pyannote audio on 2 envs :
Using python 3.10
one is macbook m2
the other one is on linux cloud, using beam.cloud
The installation seems broken on the two levels.
First, the conda command do not work on Macs, because of version compatibility.
Second the commande pip install pyannote.audio do not seems to work either.
On the cloud, when install, I can not find in Pipeline in pyannote.audio import.

I've see a chunk of code (https://huggingface.co/pyannote/segmentation) where the pipe is imported with models (Btw the first two lines of code are wrong, see importe model and usage at Model)
Am I the only one experiencing those problems?

Here is the install response on Mac :

rc/_hmmc.cpp:35:34: warning: braces around scalar initializer [-Wbraced-scalar-init] auto log = py::array_t<double>{{n}}; ^~~ src/_hmmc.cpp:63:39: warning: braces around scalar initializer [-Wbraced-scalar-init] auto scaling_ = py::array_t<double>{{ns}}; ^~~~ src/_hmmc.cpp:66:8: error: call to deleted constructor of 'pybind11::gil_scoped_release' auto nogil = py::gil_scoped_release{}; ^ ~~~~~~~~~~~~~~~~~~~~~~~~ /private/var/folders/xk/15kdb5qd7hzgslpkbw6mzvy00000gn/T/pip-install-noyfhjeh/hmmlearn_d290987eb9b947adb965e39ff85b5c02/.eggs/pybind11-2.10.4-py3.10.egg/pybind11/include/pybind11/gil.h:155:5: note: 'gil_scoped_release' has been explicitly marked deleted here gil_scoped_release(const gil_scoped_release &) = delete; ^ src/_hmmc.cpp:120:8: error: call to deleted constructor of 'pybind11::gil_scoped_release' auto nogil = py::gil_scoped_release{}; ^ ~~~~~~~~~~~~~~~~~~~~~~~~ /private/var/folders/xk/15kdb5qd7hzgslpkbw6mzvy00000gn/T/pip-install-noyfhjeh/hmmlearn_d290987eb9b947adb965e39ff85b5c02/.eggs/pybind11-2.10.4-py3.10.egg/pybind11/include/pybind11/gil.h:155:5: note: 'gil_scoped_release' has been explicitly marked deleted here gil_scoped_release(const gil_scoped_release &) = delete; ^ src/_hmmc.cpp:154:8: error: call to deleted constructor of 'pybind11::gil_scoped_release' auto nogil = py::gil_scoped_release{}; ^ ~~~~~~~~~~~~~~~~~~~~~~~~ /private/var/folders/xk/15kdb5qd7hzgslpkbw6mzvy00000gn/T/pip-install-noyfhjeh/hmmlearn_d290987eb9b947adb965e39ff85b5c02/.eggs/pybind11-2.10.4-py3.10.egg/pybind11/include/pybind11/gil.h:155:5: note: 'gil_scoped_release' has been explicitly marked deleted here gil_scoped_release(const gil_scoped_release &) = delete; ^ src/_hmmc.cpp:188:8: error: call to deleted constructor of 'pybind11::gil_scoped_release' auto nogil = py::gil_scoped_release{}; ^ ~~~~~~~~~~~~~~~~~~~~~~~~ /private/var/folders/xk/15kdb5qd7hzgslpkbw6mzvy00000gn/T/pip-install-noyfhjeh/hmmlearn_d290987eb9b947adb965e39ff85b5c02/.eggs/pybind11-2.10.4-py3.10.egg/pybind11/include/pybind11/gil.h:155:5: note: 'gil_scoped_release' has been explicitly marked deleted here gil_scoped_release(const gil_scoped_release &) = delete; ^ src/_hmmc.cpp:222:8: error: call to deleted constructor of 'pybind11::gil_scoped_release' auto nogil = py::gil_scoped_release{}; ^ ~~~~~~~~~~~~~~~~~~~~~~~~ /private/var/folders/xk/15kdb5qd7hzgslpkbw6mzvy00000gn/T/pip-install-noyfhjeh/hmmlearn_d290987eb9b947adb965e39ff85b5c02/.eggs/pybind11-2.10.4-py3.10.egg/pybind11/include/pybind11/gil.h:155:5: note: 'gil_scoped_release' has been explicitly marked deleted here gil_scoped_release(const gil_scoped_release &) = delete; ^ src/_hmmc.cpp:258:8: error: call to deleted constructor of 'pybind11::gil_scoped_release' auto nogil = py::gil_scoped_release{}; ^ ~~~~~~~~~~~~~~~~~~~~~~~~ /private/var/folders/xk/15kdb5qd7hzgslpkbw6mzvy00000gn/T/pip-install-noyfhjeh/hmmlearn_d290987eb9b947adb965e39ff85b5c02/.eggs/pybind11-2.10.4-py3.10.egg/pybind11/include/pybind11/gil.h:155:5: note: 'gil_scoped_release' has been explicitly marked deleted here gil_scoped_release(const gil_scoped_release &) = delete; ^ src/_hmmc.cpp:289:47: warning: braces around scalar initializer [-Wbraced-scalar-init] auto state_sequence_ = py::array_t<ssize_t>{{ns}}; ^~~~ src/_hmmc.cpp:293:8: error: call to deleted constructor of 'pybind11::gil_scoped_release' auto nogil = py::gil_scoped_release{}; ^ ~~~~~~~~~~~~~~~~~~~~~~~~ /private/var/folders/xk/15kdb5qd7hzgslpkbw6mzvy00000gn/T/pip-install-noyfhjeh/hmmlearn_d290987eb9b947adb965e39ff85b5c02/.eggs/pybind11-2.10.4-py3.10.egg/pybind11/include/pybind11/gil.h:155:5: note: 'gil_scoped_release' has been explicitly marked deleted here gil_scoped_release(const gil_scoped_release &) = delete; ^ 3 warnings and 7 errors generated. error: command '/usr/bin/clang' failed with exit code 1

Here is the import problem on cloud :
/usr/local/lib/python3.10/dist-packages/torchaudio/_internal/module_utils.py:99: UserWarning: Failed to import soundfile. 'soundfile' backend is not available. warnings.warn("Failed to import soundfile. 'soundfile' backend is not available.") Traceback (most recent call last): File "<stdin>", line 1, in <module> File "/usr/local/lib/python3.10/dist-packages/pyannote/audio/__init__.py", line 29, in <module> from .core.inference import Inference File "/usr/local/lib/python3.10/dist-packages/pyannote/audio/core/inference.py", line 35, in <module> from pyannote.audio.core.model import Model File "/usr/local/lib/python3.10/dist-packages/pyannote/audio/core/model.py", line 46, in <module> from pyannote.audio.core.task import Problem, Resolution, Specifications, Task File "/usr/local/lib/python3.10/dist-packages/pyannote/audio/core/task.py", line 45, in <module> from torch_audiomentations import Identity File "/usr/local/lib/python3.10/dist-packages/torch_audiomentations/__init__.py", line 1, in <module> from .augmentations.background_noise import AddBackgroundNoise File "/usr/local/lib/python3.10/dist-packages/torch_audiomentations/augmentations/background_noise.py", line 10, in <module> from ..utils.file import find_audio_files_in_paths File "/usr/local/lib/python3.10/dist-packages/torch_audiomentations/utils/file.py", line 5, in <module> import soundfile File "/usr/local/lib/python3.10/dist-packages/soundfile.py", line 142, in <module> raise OSError('sndfile library not found') OSError: sndfile library not found

The text was updated successfully, but these errors were encountered:

github-actions · 2023-03-30T13:06:38Z

Thank you for your issue. Give us a little time to review it.

PS. You might want to check the FAQ if you haven't done so already.

This is an automated reply, generated by FAQtory

hbredin · 2023-03-30T13:21:56Z

To maximise the probability of someone answering your question:

if your issue is a bug report, please provide a minimum reproducible example, e.g. a link to a self-contained Google Colab notebook (i.e. containing everthing needed to reproduce the bug: installation of pyannote.audio, downloads of models or test data, etc...)
if your issue is a feature request, please read this first and update your request accordingly.

calippo · 2023-03-31T12:03:58Z

I'm having the same issue on Python 3.10 (update downgrade to 3.8) macbook m1.

I basically cannot install hmmlearn 2.7 or 2.8 that is required by pyannote.audio (I can install 2.6 though).

[Update] I solved the problem building hmmlearn locally, but it's not optimal.

Jacubeit · 2023-03-31T20:55:26Z

Same...
Apple M1, 13.2.1
Python3.8/3.9

from pydub import AudioSegment
from pyannote.audio import Pipeline

in VS Code: Import "pyannote.audio" could not be resolved

executing the file gives the following error:

File "some_file.py", line 12, in <module>
    from pyannote.audio import Pipeline
  File "some_path/venv8/lib/python3.8/site-packages/pyannote/audio/__init__.py", line 29, in <module>
    from .core.inference import Inference
  File "some_path/venv8/lib/python3.8/site-packages/pyannote/audio/core/inference.py", line 34, in <module>
    from pyannote.audio.core.io import AudioFile
  File "some_path/venv8/lib/python3.8/site-packages/pyannote/audio/core/io.py", line 38, in <module>
    import torchaudio
  File "some_path/venv8/lib/python3.8/site-packages/torchaudio/__init__.py", line 1, in <module>
    from torchaudio import (  # noqa: F401
  File "some_path/venv8/lib/python3.8/site-packages/torchaudio/_extension.py", line 135, in <module>
    _init_extension()
  File "some_path/venv8/lib/python3.8/site-packages/torchaudio/_extension.py", line 105, in _init_extension
    _load_lib("libtorchaudio")
  File "some_path/venv8/lib/python3.8/site-packages/torchaudio/_extension.py", line 52, in _load_lib
    torch.ops.load_library(path)
  File "some_path/venv8/lib/python3.8/site-packages/torch/_ops.py", line 643, in load_library
    ctypes.CDLL(path)
  File "/opt/homebrew/Cellar/python@3.8/3.8.16/Frameworks/Python.framework/Versions/3.8/lib/python3.8/ctypes/__init__.py", line 373, in __init__
    self._handle = _dlopen(self._name, mode)
OSError: dlopen(some_path/venv8/lib/python3.8/site-packages/torchaudio/lib/libtorchaudio.so, 0x0006): Symbol not found: __ZN2at4_ops9_pad_enum4callERKNS_6TensorEN3c108ArrayRefIxEExNS5_8optionalIdEE
  Referenced from: <880489EC-8304-3310-A69F-4721C6545626> some_path/venv8/lib/python3.8/site-packages/torchaudio/lib/libtorchaudio.so
  Expected in:     <ADB29998-4917-3D35-B2A2-77968871B420> some_path/venv8/lib/python3.8/site-packages/torch/lib/libtorch_cpu.dylib

May be unrelated though.

calippo · 2023-04-01T11:14:19Z

I have the very same error after trying to overcome the installation issue.

mayeaux · 2023-04-03T11:12:35Z

This should work for the hmmlearn issue:
conda install -c conda-forge hmmlearn

also run this
conda install -c conda-forge libsndfile

Should fix the soundfile backend error

Jacubeit · 2023-04-04T08:50:30Z

what worked for me:
requirements.txt:

torch==1.13.1
torchvision==0.14.1
torchaudio==0.13.1   
git+https://github.com/hmmlearn/hmmlearn
git+https://github.com/pyannote/pyannote-audio.git@develop

remic33 · 2023-04-04T11:27:17Z

what worked for me: requirements.txt:

torch==1.13.1
torchvision==0.14.1
torchaudio==0.13.1   
git+https://github.com/hmmlearn/hmmlearn
git+https://github.com/pyannote/pyannote-audio.git@develop

Working on mac with those install thanks.
For the beam.cloud install, linux packages were missing, just have to install them using
apt-get update && apt-get install -y libsndfile1 && apt-get install ffmpeg -y
Thanks a lot.

Majdoddin · 2023-04-10T10:36:28Z

@Jacubeit Worked for me on Colab, thanks 👍

PierreVannier · 2023-04-10T16:56:16Z

I've got an M1 pro on Ventura 13.3.1 (22E261) and I've got the same error as @Jacubeit .
Traceback (most recent call last):
File "/Users/pierrevannier/whisper/whisper.cpp/speechbrain/pyannote_test.py", line 1, in <module> from pyannote.audio import Pipeline File "/Users/pierrevannier/miniforge3/envs/pyannote/lib/python3.9/site-packages/pyannote/audio/__init__.py", line 29, in <module> from .core.inference import Inference File "/Users/pierrevannier/miniforge3/envs/pyannote/lib/python3.9/site-packages/pyannote/audio/core/inference.py", line 34, in <module> from pyannote.audio.core.io import AudioFile File "/Users/pierrevannier/miniforge3/envs/pyannote/lib/python3.9/site-packages/pyannote/audio/core/io.py", line 38, in <module> import torchaudio File "/Users/pierrevannier/miniforge3/envs/pyannote/lib/python3.9/site-packages/torchaudio/__init__.py", line 1, in <module> from torchaudio import ( # noqa: F401 File "/Users/pierrevannier/miniforge3/envs/pyannote/lib/python3.9/site-packages/torchaudio/_extension.py", line 135, in <module> _init_extension() File "/Users/pierrevannier/miniforge3/envs/pyannote/lib/python3.9/site-packages/torchaudio/_extension.py", line 105, in _init_extension _load_lib("libtorchaudio") File "/Users/pierrevannier/miniforge3/envs/pyannote/lib/python3.9/site-packages/torchaudio/_extension.py", line 52, in _load_lib torch.ops.load_library(path) File "/Users/pierrevannier/miniforge3/envs/pyannote/lib/python3.9/site-packages/torch/_ops.py", line 643, in load_library ctypes.CDLL(path) File "/Users/pierrevannier/miniforge3/envs/pyannote/lib/python3.9/ctypes/__init__.py", line 374, in __init__ self._handle = _dlopen(self._name, mode) OSError: dlopen(/Users/pierrevannier/miniforge3/envs/pyannote/lib/python3.9/site-packages/torchaudio/lib/libtorchaudio.so, 0x0006): Symbol not found: __ZN2at4_ops9_pad_enum4callERKNS_6TensorEN3c108ArrayRefIxEExNS5_8optionalIdEE Referenced from: <3CEB7D0B-51E0-3459-B194-88110B2E8315> /Users/pierrevannier/miniforge3/envs/pyannote/lib/python3.9/site-packages/torchaudio/lib/libtorchaudio.so Expected in: <AE6DCE26-A528-35ED-BB3D-88890D27E6B9> /Users/pierrevannier/miniforge3/envs/pyannote/lib/python3.9/site-packages/torch/lib/libtorch_cpu.dylib

mayeaux · 2023-04-16T20:17:43Z

what worked for me: requirements.txt:

torch==1.13.1
torchvision==0.14.1
torchaudio==0.13.1   
git+https://github.com/hmmlearn/hmmlearn
git+https://github.com/pyannote/pyannote-audio.git@develop

Worked for me as well to get PyTorch working for WhisperX

MatNab007 · 2023-07-09T14:45:17Z

Hi, can anyone help me with this error?

I try to run the code
segmented, n_clusters, hyp_df, result_hypo = diarization('/content/drive/MyDrive/Speaker_Diarization/output_mono_last.wav')
print(n_clusters)
print(hyp_df)
result_hypo

but i got the error

The repository pyannote_pyannote-audio does not belong to the list of trusted repositories and as such cannot be downloaded. Do you trust this repository and wish to add it to the trusted list of repositories (y/N)?y
Downloading: "https://github.com/pyannote/pyannote-audio/zipball/main" to /root/.cache/torch/hub/main.zip

FileNotFoundError Traceback (most recent call last)
in <cell line: 2>()
1 #Give the path of audio file for Speaker Diarization (It should be Mono type.)
----> 2 segmented, n_clusters, hyp_df, result_hypo = diarization('/content/drive/MyDrive/Speaker_Diarization/output_mono_last.wav')
3 print(n_clusters)
4 print(hyp_df)
5 result_hypo

7 frames
/usr/lib/python3.10/importlib/_bootstrap_external.py in get_data(self, path)

FileNotFoundError: [Errno 2] No such file or directory: '/root/.cache/torch/hub/pyannote_pyannote-audio_main/hubconf.py'

MatNab007 · 2023-07-09T14:46:22Z

I already download what they told to do, but there is no file name hubconf.py

I cannot understand what to do

phdosilva · 2023-09-13T18:39:00Z

what worked for me: requirements.txt:

torch==1.13.1
torchvision==0.14.1
torchaudio==0.13.1   
git+https://github.com/hmmlearn/hmmlearn
git+https://github.com/pyannote/pyannote-audio.git@develop

That was not working in Google Colab and I resolved this by testing:

!pip install torch torchvision torchaudio git+https://github.com/hmmlearn/hmmlearn -qq https://github.com/pyannote/pyannote-audio/archive/refs/heads/develop.zip

Thanks anyway!

Simon-chai · 2023-09-19T09:19:29Z

This should work for the hmmlearn issue: conda install -c conda-forge hmmlearn

also run this conda install -c conda-forge libsndfile

Should fix the soundfile backend error

worked for me

hbredin added the cannot_reproduce label Mar 30, 2023

remic33 closed this as completed Apr 4, 2023

This was referenced Apr 6, 2023

Inferencing on CPU instead of GPU: why? #1289

Closed

installing on the google colab enviroment used to work, now it doesn't m-bain/whisperX#165

Open

faulty installatin, does not diarize m-bain/whisperX#166

Open

PierreVannier mentioned this issue Apr 10, 2023

Install problem on M1 #1321

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

install seems broken #1306

install seems broken #1306

remic33 commented Mar 30, 2023 •

edited

github-actions bot commented Mar 30, 2023

hbredin commented Mar 30, 2023

calippo commented Mar 31, 2023 •

edited

Jacubeit commented Mar 31, 2023 •

edited

calippo commented Apr 1, 2023

mayeaux commented Apr 3, 2023 •

edited

Jacubeit commented Apr 4, 2023

remic33 commented Apr 4, 2023

Majdoddin commented Apr 10, 2023

PierreVannier commented Apr 10, 2023

mayeaux commented Apr 16, 2023

MatNab007 commented Jul 9, 2023

MatNab007 commented Jul 9, 2023

phdosilva commented Sep 13, 2023

Simon-chai commented Sep 19, 2023

install seems broken #1306

install seems broken #1306

Comments

remic33 commented Mar 30, 2023 • edited

github-actions bot commented Mar 30, 2023

hbredin commented Mar 30, 2023

calippo commented Mar 31, 2023 • edited

Jacubeit commented Mar 31, 2023 • edited

calippo commented Apr 1, 2023

mayeaux commented Apr 3, 2023 • edited

Jacubeit commented Apr 4, 2023

remic33 commented Apr 4, 2023

Majdoddin commented Apr 10, 2023

PierreVannier commented Apr 10, 2023

mayeaux commented Apr 16, 2023

MatNab007 commented Jul 9, 2023

MatNab007 commented Jul 9, 2023

phdosilva commented Sep 13, 2023

Simon-chai commented Sep 19, 2023

remic33 commented Mar 30, 2023 •

edited

calippo commented Mar 31, 2023 •

edited

Jacubeit commented Mar 31, 2023 •

edited

mayeaux commented Apr 3, 2023 •

edited