We show that training a multi-headed self-attention-based deep network t...
The recently proposed Joint Energy-based Model (JEM) interprets
discrimi...
We present a method to remove unknown convolutive noise introduced to sp...
How important are different temporal speech modulations for speech
recog...
Conventional Frequency Domain Linear Prediction (FDLP) technique models ...
We propose a technique to compute spectrograms using Frequency Domain Li...
Wav2vec-C introduces a novel representation learning technique combining...
Quality of data plays an important role in most deep learning tasks. In ...