Segmentation Fault when loading pyannote/speaker-diarization-3.0 in rockylinux9/python3 environment

parthraghav commented 11 months ago

Hello,

I encountered a problem when trying to load the pretrained pyannote/speaker-diarization-3.0 model from the pyannote.audio pipeline in the rockylinux9/python3 environment. Specifically, I get a Segmentation Fault (SIGSEGV). Interestingly, I can load the model successfully on my Mac M1.

Environment:

	Version
Python	3.9
OS / Docker Image	rockylinux:9.2.20230513
pyannote-audio	3.0.0
torch	2.1.0
torchaudio	2.1.0

(venv) sh-5.1# pip freeze | grep pyannote

pyannote.audio==3.0.0
pyannote.core==5.0.0
pyannote.database==5.0.1
pyannote.metrics==3.2.1
pyannote.pipeline==3.0.1

Error Traceback:

  torchaudio.set_audio_backend("soundfile")
Downloading (…)lve/main/config.yaml: 100%|█████████████████████████████████████████████████████████████████████| 467/467 [00:00<00:00, 10.5kB/s]
torchvision is not available - cannot save figures
Downloading pytorch_model.bin: 100%|███████████████████████████████████████████████████████████████████████| 5.91M/5.91M [00:00<00:00, 13.7MB/s]
Downloading (…)lve/main/config.yaml: 100%|█████████████████████████████████████████████████████████████████████| 399/399 [00:00<00:00, 67.1kB/s]
Fatal Python error: Segmentation fault

Thread 0x0000ffff940e1120 (most recent call first):
  File "/usr/lib64/python3.9/threading.py", line 316 in wait
  File "/usr/lib64/python3.9/threading.py", line 581 in wait
  File "/usr/local/lib/python3.9/site-packages/tqdm/_monitor.py", line 60 in run
  File "/usr/lib64/python3.9/threading.py", line 980 in _bootstrap_inner
  File "/usr/lib64/python3.9/threading.py", line 937 in _bootstrap

Current thread 0x0000ffff969cf020 (most recent call first):
  File "/usr/local/lib/python3.9/site-packages/asteroid_filterbanks/enc_dec.py", line 216 in multishape_conv1d
  File "/usr/local/lib/python3.9/site-packages/asteroid_filterbanks/scripting.py", line 37 in wrapper
  File "/usr/local/lib/python3.9/site-packages/asteroid_filterbanks/enc_dec.py", line 177 in forward
  File "/usr/local/lib64/python3.9/site-packages/torch/nn/modules/module.py", line 1527 in _call_impl
  File "/usr/local/lib64/python3.9/site-packages/torch/nn/modules/module.py", line 1518 in _wrapped_call_impl
  File "/usr/local/lib/python3.9/site-packages/pyannote/audio/models/blocks/sincnet.py", line 87 in forward
  File "/usr/local/lib64/python3.9/site-packages/torch/nn/modules/module.py", line 1527 in _call_impl
  File "/usr/local/lib64/python3.9/site-packages/torch/nn/modules/module.py", line 1518 in _wrapped_call_impl
  File "/usr/local/lib/python3.9/site-packages/pyannote/audio/models/segmentation/PyanNet.py", line 172 in forward
  File "/usr/local/lib64/python3.9/site-packages/torch/nn/modules/module.py", line 1527 in _call_impl
  File "/usr/local/lib64/python3.9/site-packages/torch/nn/modules/module.py", line 1518 in _wrapped_call_impl
  File "/usr/local/lib/python3.9/site-packages/pyannote/audio/core/model.py", line 195 in example_output
  File "/usr/lib64/python3.9/functools.py", line 993 in __get__
  File "/usr/local/lib/python3.9/site-packages/pyannote/audio/pipelines/speaker_diarization.py", line 148 in __init__
  File "/usr/local/lib/python3.9/site-packages/pyannote/audio/core/pipeline.py", line 136 in from_pretrained
  File "/app/audio.py", line 6 in <module>
Segmentation fault

Code to Reproduce:

import faulthandler
faulthandler.enable()
from pyannote.audio import Pipeline
from shared.utils import load_env

env = load_env()
# the following line throws a segmentation fault
pipeline = Pipeline.from_pretrained("pyannote/speaker-diarization-3.0", use_auth_token=env["HF_ACCESS_TOKEN"])

To reproduce the bug:

Clone the Docker Container - https://github.com/parthraghav/pyannote-seg-fault and cd into the cloned repo.
Build the image by running docker-compose up --no-start --build.
Open the terminal in the GUI or using docker exec -it [container_id_or_name] /bin/bash cmd.
Run python3 audio.py

GDB Backtrace

(gdb) run audio.py
Starting program: /usr/bin/python3 audio.py
warning: Error disabling address space randomization: Operation not permitted
[Thread debugging using libthread_db enabled]
Using host libthread_db library "/lib64/libthread_db.so.1".
[New Thread 0xffff81407120 (LWP 234)]
[New Thread 0xffff7ebf7120 (LWP 235)]
[New Thread 0xffff7e3e7120 (LWP 236)]
[New Thread 0xffff6c455120 (LWP 237)]
[New Thread 0xffff69c45120 (LWP 238)]
[New Thread 0xffff67435120 (LWP 239)]
[Thread 0xffff67435120 (LWP 239) exited]
[Thread 0xffff69c45120 (LWP 238) exited]
[Thread 0xffff6c455120 (LWP 237) exited]
[Thread 0xffff7e3e7120 (LWP 236) exited]
[Thread 0xffff7ebf7120 (LWP 235) exited]
[Thread 0xffff81407120 (LWP 234) exited]
[Detaching after fork from child process 240]
[New Thread 0xffff7e3e7120 (LWP 241)]
[New Thread 0xffff7ebf7120 (LWP 242)]
[New Thread 0xffff81407120 (LWP 243)]
[Thread 0xffff81407120 (LWP 243) exited]
[Thread 0xffff7ebf7120 (LWP 242) exited]
[Thread 0xffff7e3e7120 (LWP 241) exited]
[Detaching after fork from child process 244]
/usr/local/lib/python3.9/site-packages/pyannote/audio/core/io.py:43: UserWarning: torchaudio._backend.set_audio_backend has been deprecated. With dispatcher enabled, this function is no-op. You can remove the function call.
  torchaudio.set_audio_backend("soundfile")
/usr/local/lib/python3.9/site-packages/torch_audiomentations/utils/io.py:27: UserWarning: torchaudio._backend.set_audio_backend has been deprecated. With dispatcher enabled, this function is no-op. You can remove the function call.
  torchaudio.set_audio_backend("soundfile")
torchvision is not available - cannot save figures
[New Thread 0xffff81407120 (LWP 245)]
[New Thread 0xffff7ebf7120 (LWP 246)]
[New Thread 0xffff7e3e7120 (LWP 247)]

Thread 1 "python3" received signal SIGSEGV, Segmentation fault.
0x0000ffff83940740 in __aarch64_cas4_acq () from /lib64/libc.so.6
(gdb) bt
#0  0x0000ffff83940740 in __aarch64_cas4_acq () from /lib64/libc.so.6
#1  0x0000ffff838c51f0 in readdir64 () from /lib64/libc.so.6
#2  0x0000ffff73a34bc4 in Xbyak_aarch64::util::Cpu::getFilePathMaxTailNumPlus1(char const*) ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#3  0x0000ffff73a34d5c in Xbyak_aarch64::util::Cpu::Cpu() () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#4  0x0000ffff73ee6130 in dnnl::impl::cpu::aarch64::jit_sve_512_1x1_conv_kernel::init_conf(dnnl::impl::cpu::aarch64::jit_1x1_conv_conf_t&, dnnl::impl::convolution_desc_t const&, dnnl::impl::memory_desc_wrapper const&, dnnl::impl::memory_desc_wrapper const&, dnnl::impl::memory_desc_wrapper const&, dnnl_primitive_attr const&, int, bool) () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#5  0x0000ffff73da0784 in dnnl_status_t dnnl::impl::primitive_desc_t::create<dnnl::impl::cpu::aarch64::jit_sve_512_1x1_convolution_fwd_t<(dnnl_data_type_t)3, (dnnl_data_type_t)3, (dnnl_data_type_t)3>::pd_t>(dnnl::impl::primitive_desc_t**, dnnl::impl::op_desc_t const*, dnnl_primitive_attr const*, dnnl_engine*, dnnl::impl::primitive_desc_t const*) () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#6  0x0000ffff731d4e5c in dnnl::impl::primitive_desc_iterator_t::operator++() [clone .isra.0] ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#7  0x0000ffff731d575c in dnnl_primitive_desc::init() () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#8  0x0000ffff731d6518 in dnnl::impl::primitive_desc_create(dnnl_primitive_desc**, dnnl_engine*, dnnl::impl::op_desc_t const*, dnnl_primitive_desc const*, dnnl_primitive_attr const*) () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#9  0x0000ffff731ac014 in dnnl_convolution_forward_primitive_desc_create ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#10 0x0000ffff6ffb84bc in dnnl::convolution_forward::primitive_desc::primitive_desc(dnnl::engine const&, dnnl::prop_kind, dnnl::algorithm, dnnl::memory::desc const&, dnnl::memory::desc const&, dnnl::memory::desc const*, dnnl::memory::desc const&, std::vector<long, std::allocator<long> > const&, std::vector<long, std::allocator<long> > const*, std::vector<long, std::allocator<long> > const&, std::vector<long, std::allocator<long> > const&, dnnl::primitive_attr const&, bool) () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#11 0x0000ffff70091824 in std::pair<dnnl::convolution_forward::primitive_desc, dnnl::convolution_forward> ideep::convolution_forward::get_primitive_desc<false>(ideep::tensor::desc const&, ideep::tensor::desc const&, ideep::tensor::desc const&, ideep::tensor::desc const&, std::vector<long, std::allocator<long> > const&, std::vector<long, std::allocator<long> > const&, std::vector<long, std::allocator<long> > const&, std::vector<long, std::allocator<long> > const&, unsigned long, bool, ideep::attr_t const&, dnnl::algorithm, dnnl::prop_kind, ideep::engine const&) [clone .isra.0] () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#12 0x0000ffff700a20a8 in void ideep::convolution_forward::compute_dispatch<false, true, true>(ideep::tensor const&, ideep::tensor const&, ideep::tensor const&, std::vector<long, std::allocator<long> > const&, ideep::tensor&, std::vector<long, std::allocator<long> > const&, std::vector<long, std::allocator<long> > const&, std::vector<long, std::allocator<long> > const&, std::vector<long, std::allocator<long> > const&, int, bool, ideep::attr_t const&, dnnl::algorithm, dnnl::prop_kind, ideep::engine const&) ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#13 0x0000ffff70094318 in at::native::_mkldnn_convolution_out(at::Tensor const&, at::Tensor const&, at::Tensor const&, std::vector<long, std::allocator<long> >&, ideep::tensor&, c10::ArrayRef<long>, c10::ArrayRef<long>, c10::ArrayRef<long>, long, bool, ideep::attr_t const&) ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#14 0x0000ffff70094c04 in at::native::_mkldnn_convolution(at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<long>, c10::ArrayRef<long>, long, bool, c10::basic_string_view<char>, c10::List<c10::optional<c10::Scalar> >, c10::optional<c10::basic_string_view<char> >) () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#15 0x0000ffff70095168 in at::native::mkldnn_convolution(at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<long>, c10::ArrayRef<long>, long) () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#16 0x0000ffff7092cdd0 in c10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoFunctor_<c10::CompileTimeFunctionPointer<at::Tensor (at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, c10::ArrayRef<long>, long), &at::(anonymous namespace)::(anonymous namespace)::wrapper_CompositeExplicitAutograd__mkldnn_convolution>, at::Tensor, c10::guts::typelist::typelist<at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, c10::ArrayRef<long>, long> >, at::Tensor (at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, c10::ArrayRef<long>, long)>::call(c10::OperatorKernel*, c10::DispatchKeySet, at::Tensor const&, at::Tensor const&, c10::optional--Type <RET> for more, q to quit, c to continue without paging--
<at::Tensor> const&, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, c10::ArrayRef<long>, long) ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#17 0x0000ffff701f9bec in at::_ops::mkldnn_convolution::call(at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, c10::ArrayRef<long>, long) () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#18 0x0000ffff6fac82c8 in at::native::_convolution(at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<long>, c10::ArrayRef<long>, bool, c10::ArrayRef<long>, long, bool, bool, bool, bool) ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#19 0x0000ffff7092dfb0 in c10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoFunctor_<c10::CompileTimeFunctionPointer<at::Tensor (at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, bool, c10::ArrayRef<c10::SymInt>, long, bool, bool, bool, bool), &at::(anonymous namespace)::(anonymous namespace)::wrapper_CompositeExplicitAutograd___convolution>, at::Tensor, c10::guts::typelist::typelist<at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, bool, c10::ArrayRef<c10::SymInt>, long, bool, bool, bool, bool> >, at::Tensor (at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, bool, c10::ArrayRef<c10::SymInt>, long, bool, bool, bool, bool)>::call(c10::OperatorKernel*, c10::DispatchKeySet, at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, bool, c10::ArrayRef<c10::SymInt>, long, bool, bool, bool, bool) () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#20 0x0000ffff701e4a2c in at::_ops::_convolution::call(at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, bool, c10::ArrayRef<c10::SymInt>, long, bool, bool, bool, bool) ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#21 0x0000ffff6fabd4ec in at::native::convolution(at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<long>, c10::ArrayRef<long>, bool, c10::ArrayRef<long>, long) ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#22 0x0000ffff7092dc58 in c10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoFunctor_<c10::CompileTimeFunctionPointer<at::Tensor (at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, bool, c10::ArrayRef<c10::SymInt>, long), &at::(anonymous namespace)::(anonymous namespace)::wrapper_CompositeExplicitAutograd__convolution>, at::Tensor, c10::guts::typelist::typelist<at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, bool, c10::ArrayRef<c10::SymInt>, long> >, at::Tensor (at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, bool, c10::ArrayRef<c10::SymInt>, long)>::call(c10::OperatorKernel*, c10::DispatchKeySet, at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, bool, c10::ArrayRef<c10::SymInt>, long) ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#23 0x0000ffff701e40c8 in at::_ops::convolution::call(at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, bool, c10::ArrayRef<c10::SymInt>, long) ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#24 0x0000ffff6fabef28 in at::native::conv1d_symint(at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, long) () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#25 0x0000ffff70a990bc in c10::impl::wrap_kernel_functor_unboxed_<c10::impl::detail::WrapFunctionIntoFunctor_<c10::CompileTimeFunctionPointer<at::Tensor (at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, long), &at::(anonymous namespace)::(anonymous namespace)::wrapper_CompositeImplicitAutograd__conv1d>, at::Tensor, c10::guts::typelist::typelist<at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, long> >, at::Tensor (at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, long)>::call(c10::OperatorKernel*, c10::DispatchKeySet, at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, long) ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
#26 0x0000ffff705fdd2c in at::_ops::conv1d::call(at::Tensor const&, at::Tensor const&, c10::optional<at::Tensor> const&, c10::ArrayRef<long>, c10::ArrayRef<c10::SymInt>, c10::ArrayRef<long>, long) () from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_cpu.so
--Type <RET> for more, q to quit, c to continue without paging--
#27 0x0000ffff76c8314c in torch::autograd::THPVariable_conv1d(_object*, _object*, _object*) ()
   from /usr/local/lib64/python3.9/site-packages/torch/lib/libtorch_python.so
#28 0x0000ffff83a928f0 in cfunction_call () from /lib64/libpython3.9.so.1.0
#29 0x0000ffff83a80ddc in _PyObject_MakeTpCall () from /lib64/libpython3.9.so.1.0
#30 0x0000ffff83a7c830 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#31 0x0000ffff83a76d10 in _PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#32 0x0000ffff83a86a78 in _PyFunction_Vectorcall () from /lib64/libpython3.9.so.1.0
#33 0x0000ffff83a9140c in PyVectorcall_Call () from /lib64/libpython3.9.so.1.0
#34 0x0000ffff83a7a438 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#35 0x0000ffff83a76d10 in _PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#36 0x0000ffff83a86a78 in _PyFunction_Vectorcall () from /lib64/libpython3.9.so.1.0
#37 0x0000ffff83a78f48 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#38 0x0000ffff83a86ce0 in function_code_fastcall () from /lib64/libpython3.9.so.1.0
#39 0x0000ffff83a90f54 in method_vectorcall () from /lib64/libpython3.9.so.1.0
#40 0x0000ffff83a7a438 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#41 0x0000ffff83a76d10 in _PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#42 0x0000ffff83a869dc in _PyFunction_Vectorcall () from /lib64/libpython3.9.so.1.0
#43 0x0000ffff83a90f54 in method_vectorcall () from /lib64/libpython3.9.so.1.0
#44 0x0000ffff83a7a438 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#45 0x0000ffff83a76d10 in _PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#46 0x0000ffff83a869dc in _PyFunction_Vectorcall () from /lib64/libpython3.9.so.1.0
#47 0x0000ffff83a80654 in _PyObject_FastCallDictTstate () from /lib64/libpython3.9.so.1.0
#48 0x0000ffff83a8eebc in _PyObject_Call_Prepend () from /lib64/libpython3.9.so.1.0
#49 0x0000ffff83b4f4f4 in slot_tp_call () from /lib64/libpython3.9.so.1.0
#50 0x0000ffff83a80d40 in _PyObject_MakeTpCall () from /lib64/libpython3.9.so.1.0
#51 0x0000ffff83a7bf6c in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#52 0x0000ffff83a86ce0 in function_code_fastcall () from /lib64/libpython3.9.so.1.0
#53 0x0000ffff83a90f54 in method_vectorcall () from /lib64/libpython3.9.so.1.0
#54 0x0000ffff83a7a438 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#55 0x0000ffff83a76d10 in _PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#56 0x0000ffff83a869dc in _PyFunction_Vectorcall () from /lib64/libpython3.9.so.1.0
#57 0x0000ffff83a90f54 in method_vectorcall () from /lib64/libpython3.9.so.1.0
#58 0x0000ffff83a7a438 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#59 0x0000ffff83a76d10 in _PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#60 0x0000ffff83a869dc in _PyFunction_Vectorcall () from /lib64/libpython3.9.so.1.0
#61 0x0000ffff83a80654 in _PyObject_FastCallDictTstate () from /lib64/libpython3.9.so.1.0
#62 0x0000ffff83a8eebc in _PyObject_Call_Prepend () from /lib64/libpython3.9.so.1.0
#63 0x0000ffff83b4f4f4 in slot_tp_call () from /lib64/libpython3.9.so.1.0
#64 0x0000ffff83a80d40 in _PyObject_MakeTpCall () from /lib64/libpython3.9.so.1.0
#65 0x0000ffff83a7c5e8 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#66 0x0000ffff83a86ce0 in function_code_fastcall () from /lib64/libpython3.9.so.1.0
#67 0x0000ffff83a90f54 in method_vectorcall () from /lib64/libpython3.9.so.1.0
#68 0x0000ffff83a7a438 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#69 0x0000ffff83a76d10 in _PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#70 0x0000ffff83a869dc in _PyFunction_Vectorcall () from /lib64/libpython3.9.so.1.0
#71 0x0000ffff83a90f54 in method_vectorcall () from /lib64/libpython3.9.so.1.0
--Type <RET> for more, q to quit, c to continue without paging--
#72 0x0000ffff83a7a438 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#73 0x0000ffff83a76d10 in _PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#74 0x0000ffff83a869dc in _PyFunction_Vectorcall () from /lib64/libpython3.9.so.1.0
#75 0x0000ffff83a80654 in _PyObject_FastCallDictTstate () from /lib64/libpython3.9.so.1.0
#76 0x0000ffff83a8eebc in _PyObject_Call_Prepend () from /lib64/libpython3.9.so.1.0
#77 0x0000ffff83b4f4f4 in slot_tp_call () from /lib64/libpython3.9.so.1.0
#78 0x0000ffff83a80d40 in _PyObject_MakeTpCall () from /lib64/libpython3.9.so.1.0
#79 0x0000ffff83a7bf6c in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#80 0x0000ffff83a86ce0 in function_code_fastcall () from /lib64/libpython3.9.so.1.0
#81 0x0000ffff83a7c004 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#82 0x0000ffff83a76d10 in _PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#83 0x0000ffff83a869dc in _PyFunction_Vectorcall () from /lib64/libpython3.9.so.1.0
#84 0x0000ffff83a860c0 in object_vacall () from /lib64/libpython3.9.so.1.0
#85 0x0000ffff83b08328 in PyObject_CallFunctionObjArgs () from /lib64/libpython3.9.so.1.0
#86 0x0000ffff83a8579c in _PyObject_GenericGetAttrWithDict () from /lib64/libpython3.9.so.1.0
#87 0x0000ffff83ad96d4 in slot_tp_getattr_hook () from /lib64/libpython3.9.so.1.0
#88 0x0000ffff83a7853c in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#89 0x0000ffff83a76d10 in _PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#90 0x0000ffff83a86a78 in _PyFunction_Vectorcall () from /lib64/libpython3.9.so.1.0
#91 0x0000ffff83a806c8 in _PyObject_FastCallDictTstate () from /lib64/libpython3.9.so.1.0
#92 0x0000ffff83a8eba0 in slot_tp_init () from /lib64/libpython3.9.so.1.0
#93 0x0000ffff83a81048 in type_call () from /lib64/libpython3.9.so.1.0
#94 0x0000ffff83a912f0 in _PyObject_Call () from /lib64/libpython3.9.so.1.0
#95 0x0000ffff83a7a438 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#96 0x0000ffff83a76d10 in _PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#97 0x0000ffff83a86a78 in _PyFunction_Vectorcall () from /lib64/libpython3.9.so.1.0
#98 0x0000ffff83a90eb0 in method_vectorcall () from /lib64/libpython3.9.so.1.0
#99 0x0000ffff83a78f48 in _PyEval_EvalFrameDefault () from /lib64/libpython3.9.so.1.0
#100 0x0000ffff83a76d10 in _PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#101 0x0000ffff83b06494 in _PyEval_EvalCodeWithName () from /lib64/libpython3.9.so.1.0
#102 0x0000ffff83b06420 in PyEval_EvalCodeEx () from /lib64/libpython3.9.so.1.0
#103 0x0000ffff83b063d0 in PyEval_EvalCode () from /lib64/libpython3.9.so.1.0
#104 0x0000ffff83b43fac in run_eval_code_obj () from /lib64/libpython3.9.so.1.0
#105 0x0000ffff83b3e568 in run_mod () from /lib64/libpython3.9.so.1.0
#106 0x0000ffff83b374cc in pyrun_file () from /lib64/libpython3.9.so.1.0
#107 0x0000ffff83b36c3c in PyRun_SimpleFileExFlags () from /lib64/libpython3.9.so.1.0
#108 0x0000ffff83b325a8 in Py_RunMain () from /lib64/libpython3.9.so.1.0
#109 0x0000ffff83af4528 in Py_BytesMain () from /lib64/libpython3.9.so.1.0
#110 0x0000ffff8383c79c in __libc_start_call_main () from /lib64/libc.so.6
#111 0x0000ffff8383c86c in __libc_start_main_impl () from /lib64/libc.so.6
#112 0x0000aaaad18a0830 in _start ()

github-actions[bot] commented 11 months ago

Thank you for your issue. We found the following entry in the FAQ which you may find helpful:

Does pyannote support streaming speaker diarization?

Feel free to close this issue if you found an answer in the FAQ.

If your issue is a feature request, please read this first and update your request accordingly, if needed.

If your issue is a bug report, please provide a minimum reproducible example as a link to a self-contained Google Colab notebook containing everthing needed to reproduce the bug:

installation
data preparation
model download
etc.

Providing an MRE will increase your chance of getting an answer from the community (either maintainers or other power users).

Companies relying on pyannote.audio in production may contact me via email regarding:

paid scientific consulting around speaker diarization and speech processing in general;
custom models and tailored features (via the local tech transfer office).

This is an automated reply, generated by FAQtory

hbredin commented 11 months ago

I am afraid I won't be able to help you as I have very litte experience with docker (for now...)

📣 To companies that have successfully deployed pyannote pipelines in production, now would be the right time to chime in and help @parthraghav.

WhitWaldo commented 11 months ago

I can't speak specifically as to why you're seeing a problem, but I can share the configuration that's been working fine in Docker for me:

Environment	Name	Version
Docker image	python:3.10-slim
pyannote.audio	3.0.1
torch	2.0.1
torchaudio	2.0.2

Using Pytorch for CUDA 11.8

davidthuman commented 11 months ago

I had the same issue.

I was able to run torch==2.1.0 and torchaudio==2.1.0 with pyannote.audio==3.0.0 on my local Mac M2. I also was originally building from python:3.11.

I down-graded to torch==2.0.1 and torchaudio==2.0.2 with the same PyAnnote Audio version and was able to call Pipeline.from_pretrained within Docker without the original seg-fault issue. I also build with python:3.10-slim, so more compatibility testing can be done.

parthraghav commented 10 months ago

Thank you @hbredin @WhitWaldo @davidthuman

After rebuilding the project requirements from scratch in the docker environment, I learned that pyannote.audio==3.0.1 appears to work in a linux cloud instance and pyannote.audio==3.0.0 works on M1. I introduced separate requirement files for both x86 and arm64 (Apple Silicon) environments.

(base) parthraghav@Parths-Mac-mini worker % cat requirements-x86.txt| grep pyannote
pyannote.audio==3.0.1
pyannote.core==5.0.0
pyannote.database==5.0.1
pyannote.metrics==3.2.1
pyannote.pipeline==3.0.1

(base) parthraghav@Parths-Mac-mini worker % cat requirements-arm64.txt| grep pyannote 
pyannote.audio==3.0.0
pyannote.core==5.0.0
pyannote.database==5.0.1
pyannote.metrics==3.2.1
pyannote.pipeline==3.0.1

pyannote / pyannote-audio

Segmentation Fault when loading pyannote/speaker-diarization-3.0 in rockylinux9/python3 environment #1499