Try using patched audiolm

# LAION-related

#*laion*

# custom-patched libs

audiolm_pytorch

audiolm-pytorch-personal-hacks.zip

# use the personal_hacks branch of my fork of the repo to get print statements, etc etc.

import urllib.request

import os

import zipfile

import subprocess

zip_file_path = "audiolm-pytorch-personal-hacks.zip"

if os.path.isfile(zip_file_path):

    replace = input("personal hacks zip already exists. replacing (as well as audiolm_pytorch library...)")

    os.remove(zip_file_path)

    shutil.rmtree("audiolm_pytorch")

urllib.request.urlretrieve("https://github.com/LWProgramming/audiolm-pytorch/archive/refs/heads/personal_hacks.zip", zip_file_path)

with zipfile.ZipFile(zip_file_path, 'r') as zip_ref:

    zip_ref.extractall("audiolm-pytorch")

# move library itself to current directory

subprocess.run(['mv', 'audiolm-pytorch/audiolm-pytorch-personal_hacks/audiolm_pytorch', '.'])

# move setup.py to current directory for requirements

subprocess.run(['mv', 'audiolm-pytorch/audiolm-pytorch-personal_hacks/setup.py', '.'])

# install requirements from the patched audiolm-pytorch directory

subprocess.run(['pip', 'install', '.'])

# remove the rest of the audiolm-pytorch directory

subprocess.run(['rm', '-rf', 'audiolm-pytorch'])