Deep learning methods haverevolutionized speech recognition, image recognition, and
natural language processing since 2010. Each of these tasks involves a single modality in …

New waves of consumer-centric applications, such as voice search and voice interaction
with mobile devices and home entertainment systems, increasingly require automatic …

An open-source Mandarin speech corpus called AISHELL-1 is released. It is by far the
largest corpus which is suitable for conducting the speech recognition research and building …

We propose to adapt deep neural network (DNN) acoustic models to a target speaker by
supplying speaker identity vectors (i-vectors) as input features to the network in parallel with …

Recurrent neural network architectures have been shown to efficiently model long term
temporal dependencies between acoustic events. However the training time of recurrent …

We present a structured overview of adaptation algorithms for neural network-based speech
recognition, considering both hybrid hidden Markov model/neural network systems and end …

This paper proposes a simple yet effective model-based neural network speaker adaptation
technique that learns speaker-specific hidden unit contributions given adaptation data …

This work presents a broad study on the adaptation of neural network acoustic models by
means of learning hidden unit contributions (LHUC)-a method that linearly re-combines …

In acoustic modeling, speaker adaptive training (SAT) has been a long-standing technique
for the traditional Gaussian mixture models (GMMs). Acoustic models trained with SAT …

Multi-style training, using data which emulates a variety of possible test scenarios, is a
popular approach towards robust acoustic modeling. However acoustic models capable of …