SPEAKER-INDEPENDENT ISOLATED DIGIT RECOGNITION - MULTILAYER PERCEPTRONS VS DYNAMIC TIME WARPING

被引：12

作者：

BOTTOU, L ^{[1
]}

SOULIE, FF ^{[1
]}

BLANCHET, P ^{[1
]}

LIENARD, JS ^{[1
]}

机构：

[1] LAB INFORMAT MEAN & SCI INGN,ORSAY,FRANCE

来源：

NEURAL NETWORKS | 1990年 / 3卷 / 04期

关键词：

Dynamic time warping; Hidden cells states clustering; Isolated digits recognition; Speaker independence; Speech recognition; Time delay neural network;

D O I：

10.1016/0893-6080(90)90028-J

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Former experiments have shown the benefit of using specific multi-layer architectures, the so-called time delay neural networks, for phoneme recognition (Waibel, Hanazawa, Hinton, Shikano, & Lang, 1988). Similar experiments on a speaker-independent task were also performed on a small set of minimal pairs (Bottou, 1988). In this paper we focus on a speaker-independent, global word recognition task with time delay networks. We first describe these networks as a way for learning feature extractors by constrained back-propagation. Such a time-delay network is shown to be capable of dealing with a near real-sized problem: French digit recognition. The results are discussed and compared, on the same data sets, with those obtained with a classical time warping system. © 1990.

引用

页码：453 / 465

页数：13