Speech recognition using recurrent neural prediction model
Speech recognition using recurrent neural prediction model
复制标题
使用循环神经预测模型的语音识别
DOI：
10.1002/scj.1194
复制
发表时间：
2000
期刊：
影响因子：
--
通讯作者：
Haruhisa Takahashi
中科院分区：
文献类型：
--
作者：
Toru Uchiyama;Haruhisa Takahashi
Iso和Watanabe提出的神经预测模型（NPM）是一种成功的语音识别神经网络，具有较高的识别率。该模型使用多层感知器进行模式预测（而不是模式识别），对非特定人孤立词的识别率高达99.8%。本文提出了一种递归神经预测模型（RNPM），并提出了该模型的递归网络结构。所提出的模型非常显着地减少了网络的大小，具有与原始模型一样高的识别率，并且对于与说话人无关的孤立词具有高的学习效率。© 2003威利期刊公司Syst Comp Jpn，34（2）：100-107，2003;在线发表于Wiley InterScience（www.interscience.wiley.com）。DOI 10.1002/scj.1194
The neural prediction model (NPM) proposed by Iso and Watanabe is a successful example of a speech recognition neural network with a high recognition rate. This model uses multilayer perceptrons for pattern prediction (not for pattern recognition), and achieves a recognition rate as high as 99.8% for speaker-independent isolated words. This paper proposes a recurrent neural prediction model (RNPM), and a recurrent network architecture for this model. The proposed model very significantly reduces the size of the network, with as high a recognition rate as the original model, and with a high efficiency of learning, for speaker-independent isolated words. © 2003 Wiley Periodicals, Inc. Syst Comp Jpn, 34(2): 100–107, 2003; Published online in Wiley InterScience (www.interscience.wiley.com). DOI 10.1002/scj.1194