【摘 要】
：
Recently long short-term memory language model(LSTMLM)has received tremendous interests from both language and speech communities,due to its superiorty on modelling long-term dependency.Moreover,integ
【机 构】
：
Key Laboratory of Shanghai Education Commission for Intelligent Interaction and Cognitive Engineerin
【出 处】
：
第十六届全国计算语言学学术会议暨第五届基于自然标注大数据的自然语言处理国际学术研讨会
论文部分内容阅读
Recently long short-term memory language model(LSTMLM)has received tremendous interests from both language and speech communities,due to its superiorty on modelling long-term dependency.Moreover,integrating auxiliary information,such as context feature,into the LSTM LM has shown improved performance in perplexity(PPL).However,improper feed of auxiliary information wont give consistent gain on word error rate(WER)in a large vocabulary continuous speech recognition(LVCSR)task.To solve this problem,a multi-view LSTM LM architecture combining a tagging model is proposed in this paper.Firstly an on-line unidirectional LSTM-RNN is built as a tagging model,which can generate word-synchronized auxiliary feature.Then the auxiliary feature from the tagging model is combined with the word sequence to train a multi-view unidirectional LSTM LM.Different training modes for the tagging model and language model are explored and compared.The new architecture is evaluated on PTB,Fisher English and SMS Chinese data sets,and the results show that not only LM PPL promotion is observed,but also the improvements can be well transferred to WER reduction in ASR-rescore task.
其他文献
Local community detection is an important research focus in social network analysis.Most existing methods share the intrinsic limitation of utiliz-ing undirected and unweighted networks.In this paper,
随着互联网的发展及硬件的更新,神经网络模型被广泛应用于自然语言处理、图像识别等领域.目前,结合传统自然语言处理方法和神经网络模型正日益成为研究的热点.引入先验知识代表了传统方法的惯例,然而它们对基于神经网络模型的自然语言处理任务的影响尚不清楚.鉴于此,本文尝试探究语言层先验知识对基于神经网络模型的若干自然语言处理任务的影响.根据不同任务的特点,比较了不同先验知识和不同输入位置对不同神经网络模型的影
We take the generation of Chinese classical poetry as a sequence-to-sequence learning problem,and investigate the suitability of recurrent neural network(RNN)for poetry generation task by various qual
Understanding chemical-disease relations(CDR)from biomedicalliterature is important for biomedical research and chemical discovery.This pa-per uses a k-max pooling convolutional neural network(CNN)to
Most state-of-the-art models for named entity recognition(NER)rely on recurrent neural networks(RNNs),in particular long short-term memory(LSTM).Those models learn local and global fea-tures automatic
Word deletion(WD)errors can lead to poor comprehension of the meaning of source translated sentences in phrase-based statistical machine translation(SMT),and have a critical impact on the adequacy of
Answer selection is a crucial subtask of the open domain question answering problem.In this paper,we introduce the Bi-directional Gated Memory Network(BGMN)to model the interactions between question a
In the last decades,named entity recognition has been extensivelystudied with various supervised learning approaches depend on massive labeled data.In this paper,we focus on person name recognition in
Enabling a computer to understand a document so that itcan answer comprehension questions is a central,yet unsolved goal of Natural Language Processing,so reading comprehension of text is an important
Generating textual entailment(GTE)is a recently proposed task to study how to infer a sentence from a given premise.Current sequence-to-se-quence GTE models are prone to produce invalid sentences when