Recently, the speech community is seeing a significant trend of moving from deep neural
network based hybrid modeling to end-to-end (E2E) modeling for automatic speech …

The polarization of world languages is becoming more and more obvious. Many languages,
mainly endangered languages, are of low-resource attribute due to lack of information. Both …

Sequence-to-sequence (seq2seq) approach for low-resource ASR is a relatively new
direction in speech research. The approach benefits by performing model training without …

The goal of this paper is to develop state-of-the-art models for lip reading--visual speech
recognition. We develop three architectures and compare their accuracy and training …

Multi-media communications facilitate global interaction among people. However, despite
researchers exploring cross-lingual translation techniques such as machine translation and …

In this paper, we propose an attention-based end-to-end neural approach for small-footprint
keyword spotting (KWS), which aims to simplify the pipelines of building a production-quality …

We develop streaming keyword spotting systems using a recurrent neural network
transducer (RNN-T) model: an all-neural, end-to-end trained, sequence-to-sequence model …

The goal of this work is to automatically determine whether and when a word of interest is
spoken by a talking face, with or without the audio. We propose a zero-shot method suitable …

Training Automatic Speech Recognition (ASR) models under federated learning (FL)
settings has attracted a lot of attention recently. However, the FL scenarios often presented …

Multilingual Automated Speech Recognition (ASR) systems allow for the joint training of
data-rich and data-scarce languages in a single model. This enables data and parameter …