virtual conference icassp 2021

A Comparison Of Methods For Oov-Word Recognition On A New Public Dataset

Graphspeech: Syntax-Aware Graph Attention Network For Neural Speech Synthesis

Do As I Mean, Not As I Say: Sequence Loss Training For Spoken Language Understanding

Noise Level Limited Sub-Modeling For Diffusion Probabilistic Vocoders

Dbnet: Doa-Driven Beamforming Network For End-To-End Reverberant Sound Source Separation

Convolutional Dropout And Wordpiece Augmentation For End-To-End Speech Recognition

St-Bert: Cross-Modal Language Model Pre-Training For End-To-End Spoken Language Understanding

Cdpam: Contrastive Learning For Perceptual Audio Similarity

Stylemelgan: An Efficient High-Fidelity Adversarial Vocoder With Temporal Adaptive Normalization

End2End Acoustic To Semantic Transduction

Unlimited Sensing: An Invitation To Modulo Sampling And Painless Hdr Reconstruction Beyond Shannon

Tiny Transducer: A Highly-Efficient Speech Recognition Model On Edge Devices