We investigate the impact of aggressive low-precision representations of weights and
activations in two families of large LSTM-based architectures for Automatic Speech …

One of the significant features of extreme learning machines (ELMs) is their fast
convergence. However, in the big data environment, the ELM based on the Moore–Penrose …

State-of-the-art language models (LMs) represented by long-short term memory recurrent
neural networks (LSTM-RNNs) and Transformers are becoming increasingly complex and …

Recently, speaker verification systems benefit from deep neural networks and the size of
speaker embedding encoder increases with these sophisticated architectures. Nevertheless …

State-of-the-art neural language models represented by Transformers are becoming
increasingly complex and expensive for practical applications. Low-bit deep neural network …

Recognition of overlapped speech has been a highly challenging task to date. State-of-the-
art multi-channel speech separation system are becoming increasingly complex and …

State of the art time automatic speech recognition (ASR) systems are becoming increasingly
complex and expensive for practical applications. This paper presents the development of a …

The next crucial step in artificial intelligence involves integrating neural network models into
embedded and mobile systems. This requires designing compact and energy-efficient …

With the continuous development of deep neural networks (DNN) in recent years, the
performance of speaker verification systems has been significantly improved with the …

摘要随着电商行业的蓬勃发展, 网上购物逐渐取代线下商店成为最受欢迎的购物方式之一.
因此从海量的商品评价中挖掘出有用的信息, 对顾客购买商品和商家提高服务质量具有重要的 …