Releases: marl/openl3
v0.2.0
Release version 0.2.0 of OpenL3.
- Update embedding models with ones that have been trained with the kapre bug fixed.
- Allow loaded models to be passed in and used in process_file and get_embedding.
- Rename get_embedding_model to load_embedding_model.
v0.2.0rc0
Release candidate 0 of version 0.2.0
- Update embedding models with ones that have been trained with the kapre bug fixed.
- Allow loaded models to be passed in and used in process_file and get_embedding.
- Rename get_embedding_model to load_embedding_model.
v0.1.1
Release of v0.1.1 of OpenL3.
Update kapre to fix issue with dynamic range normalization for decibel computation when computing spectrograms.
v0.1.1rc1
Release candidate 1 of version 0.1.1
v0.1.1rc0
Release candidate 0 of version 0.1.1
Update kapre to fix issue with dynamic range normalization for decibel computation when computing spectrograms.
v0.1.0
First release of OpenL3.
Implements audio embedding models, basic extraction and file processing API, and CLI.
v0.1.0rc6
Release candidate 6 of version 0.1.0
Update docs with references and make spelling of "OpenL3" consistent.
v0.1.0rc5
Release candidate 5 of version 0.1.0
v0.1.0rc4
Release candidate 4 of version 0.1.0
This release also updates the PyPI keywords, and moves the model files directly into the module directory (instead of creating a subdirectory) to make the pip installation process easier when installing with PyPI.
v0.1.0rc3
Release candidate 3 of version 0.1.0