I need to install rubberband on a linux cluster (my home directory) and use it in an audio processing project using PyTorch. Obviously, I don't have root access and can't use "sudo apt-get", so I have to install all packages locally and on virtual environment (for python packages). The documentation/readme of rubbeband is simply too incomplete and not very helpful.
Rubberband depends on libsamplerate and libsndfile. Installation/make of these packages are easy and well explained in their mirror. I used the following classic sequence and it worked fine:
./autogen.sh
./configure --prefix=~/opt
make
make install
After installing these packages, I updated environmental variables as
export PATH=$PATH:$HOME/opt/bin
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:$HOME/opt/lib
export INCLUDE=$INCLUDE:$HOME/opt/include
The problem is the installation of rubberband. In its documentation, the following has been recommended:
$ make -f otherbuilds/Makefile.linux
But it doesn't work. I know PyRubberband is the python wrapper for rubberband; I installed it using pip, but it doesn't work. It throws the following exception:
RuntimeError: Failed to execute rubberband. Please verify that rubberband-cli is installed.
My question is how can I make (install) rubberband from source on a local directory (called /opt) and make it findable by pyrubberband (or all other packages they might need it)?
Please be specific and give details/step by step instructions (e.g., make process, where, when and how I should update environment variables like PATH, etc.).
Your time and answer are really appreciated.
PS. I have already checked this post and it does not answer my question.
