This repository is providing a compiled version of aubio-python, the Python module for aubio, which is not installed with the installation of the aubio package that's in the official Termux repository.
This program was entirely built within Termux on my ARMv8 (aarch64/arm64) phone.
After many attempts of trying to install from pip didn't bear fruit, I've tried using proot-distro environments (which worked with pip), but my goal was to have it working in a vanilla Termux environment. This build is based on release (v.0.5.0a0) from the Git repo, instead of the one from pypi's v.0.4.9 one.
It is pretty easy: just extract the tar.gz archive, but make a folder first to unpack.
Assuming that you're in the home folder, make a new directory using mkdir yourfoldername, then for extraction you can use: tar -xf aubio-python-aarch64.tar.gz -C ./yourfoldername.
The extracted folder matches the layout of the Termux prefix for ease of use.
To make everything easy, it is recommended to use mc (Midnight Commander) to copy over the files, just remember to not go
back from the com.termux/files folder (which has /home and /usr in it).
After copying the files to their own directories, you can try launching aubio by simply typing ./aubio, which would launch the program.
- If you see a message from
tarsaying that... not found in archive, use either the-C ./foldernameor--directory /path/to/folderoption to extract to a folder. - If you see a message saying
ModuleNotFoundError: No module named aubio, investigate ifaubioandaubiocuthas "executable" permissions, you can test it by going to/usr/binthen invoking./aubio. Ifbash: aubio : Permission deniedis the message you get, usechmod +x ./aubioand that should fix the issue.
If you're still experiencing the same message, go to /usr/lib/python3.11/site-packages/aubio and repeat chmod +x on all the .py files and the .so file in the directory.
If all done, you should have proper access to the program. However, note that you still need to source numpy (dependency for aubio) from either the official Termux repo or
installing numpy from pip using MATHLIB="m" pip install numpy.