keunwoochoi/kapre
A Keras extension providing GPU-accelerated audio preprocessing layers like STFT and melspectrogram for deep learning models.
Not currently ranked — collecting fresh signals.
star history
Kapre provides Keras layers that compute audio signal processing operations (STFT, ISTFT, mel-spectrogram) on GPU in real time during model execution. These layers can be inserted as the first layer of a neural network to handle audio preprocessing within the model graph, eliminating precomputation dependencies and enabling end-to-end training. It is tested against Librosa for correctness and supports both channels_first and channels_last data formats with TensorFlow 2.16+.
Frequently asked
- What is keunwoochoi/kapre?
- A Keras extension providing GPU-accelerated audio preprocessing layers like STFT and melspectrogram for deep learning models.
- Is kapre open source?
- Yes — keunwoochoi/kapre is open source, released under the MIT license.
- What language is kapre written in?
- keunwoochoi/kapre is primarily written in Python.
- How popular is kapre?
- keunwoochoi/kapre has 947 stars on GitHub.
- Where can I find kapre?
- keunwoochoi/kapre is on GitHub at https://github.com/keunwoochoi/kapre.