no code implementations • 1 Jun 2023 • Ashutosh Chaubey, Sparsh Sinha, Susmita Ghose
Speaker identification systems are deployed in diverse environments, often different from the lab conditions on which they are trained and tested.
no code implementations • 31 Mar 2022 • Ashutosh Chaubey, Sparsh Sinha, Susmita Ghose
The use of relation networks facilitates joint training of the frontend speaker encoder and the backend model.
no code implementations • 1 Oct 2021 • Sumit Shekhar, Bhanu Prakash Reddy Guda, Ashutosh Chaubey, Ishan Jindal, Avneet Jain
We propose novel rewards to account for class imbalance and user feedback in the annotation interface, to improve the active learning method.
no code implementations • 16 May 2020 • Ashutosh Chaubey, Nikhil Agrawal, Kavya Barnwal, Keerat K. Guliani, Pramod Mehta
We also cover the applications of such universal perturbations in various deep learning tasks.