Proceedings ArticleDOI
Rescoring under fuzzy measures with a multilayer neural network in a rule-based speech recognition system
O. Oppizzi,R. Quelavoine +1 more
- Vol. 3, pp 1723-1726
Reads0
Chats0
TLDR
A speech rescoring system is developed on a set of phonetic hypotheses produced by a bottom-up knowledge-based decoder in order to detect how these rules characterize different phonemes and then give a weight to each rule.Abstract:
A speech rescoring system is developed on a set of phonetic hypotheses produced by a bottom-up knowledge-based decoder An original method to automatically compute a fuzzy membership function from top-down acoustic rules statistics is compared with a possibilistic measure To aggregate the fuzzy degrees into a phonetic score, a multilayer neural network is trained on the results of all the rules in order to detect how these rules characterize different phonemes and then in order to give a weight to each rule The rescoring performance of top-down rules for fricatives is discussed on an isolated-word speech database of French with 1000 utterances pronounced by five speakersread more
Citations
More filters
Patent
Method and device for recognising a phonetic sound sequence or character sequence
TL;DR: In this paper, a method for recognising a phonetic sound sequence or a character sequence, e.g., according to the ASCII standard, is presented. But the method is limited to the case where the character sequence has a defined statement content.
Book ChapterDOI
Learning Verification in Multilayer Neural Networks
Régis Quélavoine,Pascal Nocera +1 more
TL;DR: This paper addresses the difficult problem of the learning verification in multilayer neural networks by finding the activation/inhibition power of each input feature and finding the minimal recognized patterns as long as to evaluate the robustness of the system.
References
More filters
Proceedings ArticleDOI
Explicit representation of knowledge acquired from plant historical data using neural network
Kenji Baba,I. Enbutu,M. Yoda +2 more
TL;DR: A causal index, which translates implicit knowledge contained in a neural network into an explicit representation, is proposed and a backpropagation-based learning algorithm which suppresses the nondominant causal relationships to improve association accuracy is developed using the index.
Journal ArticleDOI
A hybrid segmental neural net/hidden Markov model system for continuous speech recognition
TL;DR: A hybrid system that combines the advantages of neural networks and HMM using a multiple hypothesis (or N-best) paradigm, where the connectionist component of the system, the segmental neural net (SNN), models all the frames of a phonetic segment simultaneously, thus overcoming the well-known conditional-independence limitation of the HMM.
Proceedings ArticleDOI
Multilayer perceptrons and data analysis
Gallinari,Thiria,Soulie +2 more
TL;DR: Results are presented which permit comparison of classification tasks of multilayer perceptrons with discriminant analysis and simulations of both approaches demonstrate that multilayers perceptron with nonlinear elements outperform discriminantAnalysis.
Proceedings ArticleDOI
Diminishing the number of nodes in multi-layered neural networks
P. Nocera,R. Quelavoine +1 more
TL;DR: Two ways for diminishing the size of a multilayered neural network trained to recognise French vowels are proposed: the study of the variation of the outputs of each node and the examination of the connecting weights between the input nodes.