执行 make TH_VERSION=1.10.1 CUDA_VERSION=11.3，然后报错cython无法识别np.ndarray

[MISSLI-XIAOYU]

Describe the issue 执行 make TH_VERSION=1.10.1 CUDA_VERSION=11.3，然后报错cython， Error compiling Cython file: ------------------------------------------------------------ ... import logging import numpy as np cimport numpy as np

  def cython_fill_table(np.ndarray[np.float32_t, ndim=2] table,
                        ^
  ------------------------------------------------------------
  
  ctc_segmentation/ctc_segmentation_dyn.pyx:20:22: 'ndarray' is not a type identifier

Show the check_install.py status by using the following command

cd <espnet-root>/tools
. ./activate_python.sh; python3 check_install.py

Basic environments:

• OS information: Linux 6.8.0-59-generic #61~22.04.1-Ubuntu SMP PREEMPT_DYNAMIC Tue Apr 15 17:03:15 UTC 2 x86_64
• python version: 3.8.20
• espnet version:最新的 2025-05-21下载
• Git hash [e.g. b88e89fc7246fed4c2842b55baba884fe1b4ecc2]
  • Commit date [e.g. Tu-e Sep 1 09:32:54 2020 -0400]
• pytorch version 1.10.1

To Reproduce Steps to reproduce the behavior by showing us the specific installation commands with their arguments, e.g.,

cd <espnet-root>/tools
make TH_VERSION=1.3.1

Error logs Paste the error logs. If applicable, add screenshots to help explain your problem.

[shivamtiwari2112]

I am also not able to install ESPnet by using the given document. i am getting this error: make: *** [Makefile:282: check_install] Error 1

[sw005320]

We also recognize that the installation is unstable or broken (possibly due to numpy 2.X). We are investigating the issue, but it would be great if you could find some causes.

[shivamtiwari2112]

We also recognize that the installation is unstable or broken (possibly due to numpy 2.X). We are investigating the issue, but it would be great if you could find some causes.

Sir, I think numpy or numba is the possible reason. Because some time error is showing numpy._core.multiarray failed to import. In this installation guide, it is mentioned that Ubuntu 18 is supported. Is it also possible that it will not work in Ubuntu >=18?

[albertz]

Downgrading to ctc_segmentation-1.6.6 should fix this (that's still compatible with Numpy 1).

I think ctc_segmentation-1.7.4 which is used here needs Numpy 2.

[github-actions[bot]]

This issue is stale because it has been open for 45 days with no activity. It will be closed if no further activity occurs. Thank you for your contributions.

[github-actions[bot]]

This issue is closed. Please re-open if needed.