Issue Description
Python 3.13 removes the aifc and sunau modules that librosa depends on through audioread. This causes the application to fail with errors like the one reported in issue #1 (ModuleNotFoundError: No module named 'audioop').
Related Issues
Current Status
The pyproject.toml file already includes a Python version constraint (requires-python = ">=3.11,<3.13") to prevent installation on Python 3.13+, but we should develop a more forward-compatible solution.
Proposed Strategies
-
Wait for upstream fix: Monitor the audioread/librosa projects for updates that address Python 3.13 compatibility.
-
Fork and patch: Create a temporary fork of audioread with patches to work around the missing modules until the upstream is fixed.
-
Alternative audio processing: Investigate alternative libraries that don't rely on the removed modules:
-
Conditional imports: Modify the code to conditionally use different audio processing libraries based on the Python version.
Next Steps
I suggest investigating option 3 first, as we already have pydub and soundfile as dependencies. We could potentially refactor our audio processing code to use these libraries directly instead of relying on librosa for the affected functionality.
Issue Description
Python 3.13 removes the
aifcandsunaumodules thatlibrosadepends on throughaudioread. This causes the application to fail with errors like the one reported in issue #1 (ModuleNotFoundError: No module named 'audioop').Related Issues
aifcmodule is deprecated (and removed in Python 3.13) beetbox/audioread#144Current Status
The
pyproject.tomlfile already includes a Python version constraint (requires-python = ">=3.11,<3.13") to prevent installation on Python 3.13+, but we should develop a more forward-compatible solution.Proposed Strategies
Wait for upstream fix: Monitor the audioread/librosa projects for updates that address Python 3.13 compatibility.
Fork and patch: Create a temporary fork of audioread with patches to work around the missing modules until the upstream is fixed.
Alternative audio processing: Investigate alternative libraries that don't rely on the removed modules:
Conditional imports: Modify the code to conditionally use different audio processing libraries based on the Python version.
Next Steps
I suggest investigating option 3 first, as we already have pydub and soundfile as dependencies. We could potentially refactor our audio processing code to use these libraries directly instead of relying on librosa for the affected functionality.