Large-vocabulary isolated word recognition with fast coarse time alignment

Aktas, Abdulmesih; Kämmerer, Bernhard R.; Küpper, Wolfgang A.; Lagger, H.

doi:10.1109/icassp.1986.1169201

Cited by 5 publications

(1 citation statement)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Each word is now represented by a set of feature vectors which forms a spectral pattern of different size depending on the word length. In order to arrive at standard-size patterns, a quasi-linear division of the word pattern into 16 segments is performed [16]. The outer segments are enlarged in order to allow for imprecisely determined word boundaries and the larger variability of speech at the onset and end of a word.…”

Section: Signal Analysis and Template Generationmentioning

confidence: 99%

Design of hierarchical perceptron structures and their application to the task of isolated-word recognition

Kammerer

Kupper

1989

International Joint Conference on Neural Networks

View full text Add to dashboard Cite

Several design strategies for feed-forward networks are examined within the scope o f pattern classification. Singleand two-layer perceptron models are adapted for experiments in isolated-word recognition. Direct (one-step) classification as well as several hierarchical (two-step) schemes have been considered. For a vocabulary o f twenty English words spoken repeatedly by eleven speakers, the word classes are found to be separable by hyperplanes in the chosen feature space. Since for speaker-dependent word recognition the underlying data base contains only a small training set, an automatic expansion o f the training material improves the eneralization properties o f the networks. This methoc?accounh for a wide variety o f observable temporal structures for each word and gives a better overall estimate of the network parameters which leads to a recognition rate o f 99.5 %. For speaker-independent word reco nition, a hierarchical structure with pairwise training of?wo-class models is superior to a single uniform network (98 % average recognition rate).

show abstract

Section: Signal Analysis and Template Generationmentioning

confidence: 99%