ICASSP 2021

Extending Parrotron: An End-To-End, Speech Conversion And Speech Recognition Model For Atypical Speech

Improved Data Selection For Domain Adaptation In Asr

Non-Autoregressive Sequence-To-Sequence Voice Conversion

Divide And Conquer: One-Bit Mimo-Ofdm Detection By Inexact Expectation Maximization

Polynomial Matrix Eigenvalue Decomposition Of Spherical Harmonics For Speech Enhancement

Low-Latency Polar Decoder Using Overlapped Scl Processing

Sparse Recovery Beamforming And Upscaling In The Ray Space

Supervised Direct-Path Relative Transfer Function Learning For Binaural Sound Source Localization

Coexistence Of Communications And Cognitive Mimo Radar

Refining Automatic Speech Recognition System For Older Adults

Fast Dctts: Efficient Deep Convolutional Text-To-Speech

Ppg-Based Singing Voice Conversion With Adversarial Representation Learning