The proposed method for pruning S-Edge based State Space Models.
© Matthias Bittner

Pruning State Space Models with Model Order Reduction for Efficient Raw Audio Classification at EUSIPCO’25

The discrete recurrent forward path of our proposed S-Edge compared to the Original S5.
© Matthias Bittner, Daniel Schnöll

Efficient and interpretable raw audio classification with diagonal state space models at ECMLPKDD’25

 

 Logo von Konferenz

Diagonal State Space Models for efficient MIDI-to-RAW Audio Synthesis at DAFx25

ogo der date 2023

The paper 'Multispectral Feature Fusion for Deep Object Detection on Embedded NVIDIA Platforms' got accepted at the DATE 2023 conference, 17-19 April…

Tiny YoloV3 Power Visualization

Evaluating a neural network in terms of accuracy is widely supported in all machine learning frameworks. The power draw which is an important aspect…