Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Sebastian Ewert

Seq-U-Net: A One-Dimensional Causal U-Net for Efficient Sequence Modelling


Nov 14, 2019
Daniel Stoller, Mi Tian, Sebastian Ewert, Simon Dixon

* Code available at https://github.com/f90/Seq-U-Net 

  Access Paper or Ask Questions

Training Generative Adversarial Networks from Incomplete Observations using Factorised Discriminators


May 29, 2019
Daniel Stoller, Sebastian Ewert, Simon Dixon

* 10 pages plus 14 pages appendix. Under review. Implementation available at https://github.com/f90/FactorGAN 

  Access Paper or Ask Questions

End-to-end Lyrics Alignment for Polyphonic Music Using an Audio-to-Character Recognition Model


Feb 18, 2019
Daniel Stoller, Simon Durand, Sebastian Ewert

* 5 pages (1 for references), 2 figures, 2 tables. Camera-ready version, accepted at the International Conference on Acoustics, Speech, and Signal Processing 2019 (ICASSP) 

  Access Paper or Ask Questions

Wave-U-Net: A Multi-Scale Neural Network for End-to-End Audio Source Separation


Jun 08, 2018
Daniel Stoller, Sebastian Ewert, Simon Dixon

* 19th International Society for Music Information Retrieval Conference (ISMIR 2018) 
* 7 pages (1 for references), 4 figures, 3 tables. Appearing in the proceedings of the 19th International Society for Music Information Retrieval Conference (ISMIR 2018) (camera-ready version). Implementation available at https://github.com/f90/Wave-U-Net 

  Access Paper or Ask Questions

Adversarial Semi-Supervised Audio Source Separation applied to Singing Voice Extraction


Apr 06, 2018
Daniel Stoller, Sebastian Ewert, Simon Dixon

* 5 pages, 2 figures, 1 table. Final version of manuscript accepted for 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). Implementation available at https://github.com/f90/AdversarialAudioSeparation 

  Access Paper or Ask Questions

Jointly Detecting and Separating Singing Voice: A Multi-Task Approach


Apr 05, 2018
Daniel Stoller, Sebastian Ewert, Simon Dixon

* 10 pages, 2 figures, accepted for the 14th International Conference on Latent Variable Analysis and Signal Separation 

  Access Paper or Ask Questions

An Augmented Lagrangian Method for Piano Transcription using Equal Loudness Thresholding and LSTM-based Decoding


Jul 30, 2017
Sebastian Ewert, Mark B. Sandler

* Proceedings of the IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA), New Paltz, NY, USA, pp. 146-150, 2017 

  Access Paper or Ask Questions

Structured Dropout for Weak Label and Multi-Instance Learning and Its Application to Score-Informed Source Separation


Dec 26, 2016
Sebastian Ewert, Mark B. Sandler

* Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing (ICASSP), New Orleans, USA, pp. 2277-2281, 2017 

  Access Paper or Ask Questions