no code implementations • 8 Feb 2024 • Karim Helwani, Masahito Togami, Paris Smaragdis, Michael M. Goodwin
In this paper, we present a hybrid classical digital signal processing/deep neural network (DSP/DNN) approach to source separation (SS) highlighting the theoretical link between variational autoencoder and classical approaches to SS.
no code implementations • 1 Feb 2024 • Masahito Togami, Jean-Marc Valin, Karim Helwani, Ritwik Giri, Umut Isik, Michael M. Goodwin
The algorithm runs in real-time on 10-ms frames with a 40 ms of look-ahead.
no code implementations • 10 Oct 2023 • Karim Helwani, Erfan Soltanmohammadi, Michael M. Goodwin
Improving the interpretability of deep neural networks has recently gained increased attention, especially when the power of deep learning is leveraged to solve problems in physics.
no code implementations • 25 Sep 2023 • Jan Büthe, Ahmed Mustafa, Jean-Marc Valin, Karim Helwani, Michael M. Goodwin
Speech codec enhancement methods are designed to remove distortions added by speech codecs.
no code implementations • 29 May 2023 • Emmanouil Theodosis, Karim Helwani, Demba Ba
Employing equivariance in neural networks leads to greater parameter efficiency and improved generalization performance through the encoding of domain knowledge in the architecture; however, the majority of existing approaches require an a priori specification of the desired symmetries.
no code implementations • 3 Feb 2022 • Wo Jae Lee, Karim Helwani, Arvindh Krishnaswamy, Srikanth Tenneti
The presented approach doesn't assume the presence of labeled anomalies in the training dataset and uses a novel deep neural network architecture to learn the temporal dynamics of the multivariate time series at multiple resolutions while being robust to contaminations in the training dataset.
no code implementations • 9 Feb 2021 • Turab Iqbal, Karim Helwani, Arvindh Krishnaswamy, Wenwu Wang
For tasks such as classification, there is a good case for learning representations of the data that are invariant to such transformations, yet this is not explicitly enforced by classification losses such as the cross-entropy loss.
no code implementations • 11 Aug 2020 • Umut Isik, Ritwik Giri, Neerad Phansalkar, Jean-Marc Valin, Karim Helwani, Arvindh Krishnaswamy
Neural network applications generally benefit from larger-sized models, but for current speech enhancement models, larger scale networks often suffer from decreased robustness to the variety of real-world use cases beyond what is encountered in training data.